Updated on 2022/05/17

写真a

 
MATSUSHIMA, Toshiyasu
 
Affiliation
Faculty of Science and Engineering, School of Fundamental Science and Engineering
Job title
Professor

Concurrent Post

  • Faculty of Science and Engineering   Graduate School of Fundamental Science and Engineering

Research Institute

  • 2021
    -
    2022

    大学総合研究センター   兼任センター員

  • 2020
    -
    2022

    理工学術院総合研究所   兼任研究員

Education

  •  
    -
    1991

    Waseda University   Graduate School, Division of Science and Engineering  

  •  
    -
    1991

    Waseda University   Graduate School, Division of Science and Engineering  

  •  
    -
    1978

    Waseda University   Faculty of Science and Engineering  

Degree

  • 早稲田大学   博士(工学)

  • Engineering

Professional Memberships

  •  
     
     

    日本応用数理学会

  •  
     
     

    応用統計学会

  •  
     
     

    日本OR学会

  •  
     
     

    品質管理学会

  •  
     
     

    人工知能学会

  •  
     
     

    情報処理学会

  •  
     
     

    情報理論とその応用学会

  •  
     
     

    情報理論とその応用学会

  •  
     
     

    IEEE

  •  
     
     

    電子情報通信学会

  •  
     
     

    電子情報通信学会

  •  
     
     

    電子情報通信学会

  •  
     
     

    電子情報通信学会

▼display all

 

Research Areas

  • Applied mathematics and statistics

  • Basic mathematics

  • Information security

  • Control and system engineering

  • Control and system engineering

  • Intelligent informatics

  • Statistical science

  • Theory of informatics

  • Communication and network engineering

▼display all

Research Interests

  • 情報理論、符号理論、統計科学、学習理論、知能情報、データ科学,情報セキュリティ

Papers

  • Meta-Tree Random Forest: Probabilistic Data-Generative Model and Bayes Optimal Prediction

    Nao Dobashi, Shota Saito, Yuta Nakahara, Toshiyasu Matsushima

    Entropy   23 ( 6 ) 768 - 768  2021.06

     View Summary

    This paper deals with a prediction problem of a new targeting variable corresponding to a new explanatory variable given a training dataset. To predict the targeting variable, we consider a model tree, which is used to represent a conditional probabilistic structure of a targeting variable given an explanatory variable, and discuss statistical optimality for prediction based on the Bayes decision theory. The optimal prediction based on the Bayes decision theory is given by weighting all the model trees in the model tree candidate set, where the model tree candidate set is a set of model trees in which the true model tree is assumed to be included. Because the number of all the model trees in the model tree candidate set increases exponentially according to the maximum depth of model trees, the computational complexity of weighting them increases exponentially according to the maximum depth of model trees. To solve this issue, we introduce a notion of meta-tree and propose an algorithm called MTRF (Meta-Tree Random Forest) by using multiple meta-trees. Theoretical and experimental analyses of the MTRF show the superiority of the MTRF to previous decision tree-based algorithms.

    DOI

  • Bayes code for two-dimensional auto-regressive hidden Markov model and its application to lossless image compression

    Yuta Nakahara, Toshiyasu Matsushima

    International Workshop on Advanced Imaging Technology (IWAIT) 2020    2020.06

    DOI

  • A Stochastic Model of Block Segmentation Based on the Quadtree and the Bayes Code for It

    Yuta Nakahara, Toshiyasu Matsushima

    2020 Data Compression Conference (DCC)    2020.03

    DOI

  • A Note on Mixed Level Experimental Designs Using Augmented Orthogonal Arrays

    Junki Yamaguchi, Koki Kazama, Akira Kamatsuka, Shota Saito, Toshiyasu Matsushima

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   J103-A ( 1 )  2020.01  [Refereed]

  • Reducing the Computational and Communication Complexity of a Distributed Optimization for Regularized Logistic Regression

    Nozomi Miya, Hideyuki Masui, Hajime Jinushi, Toshiyasu Matsushima

    Proceedings of 2019 IEEE International Conference on Systems, Man, and Cybernetics     3434 - 3439  2019.10  [Refereed]

  • Model Selection of Bayesian Hierarchical Mixture of Experts Based on Variational Inference

    Yuji Iikubo, Shunsuke Horii, Toshiyasu Matsushima

    2019 IEEE International Conference on Systems, Man, and Cybernetics (SMC2019)    2019.10  [Refereed]

  • Covariance Evolution for Spatially ``Mt. Fuji'' Coupled LDPC Codes

    Yuta Nakahara, Toshiyasu Matsushima

    IEEE Information Theory Workshop (ITW) 2019    2019.08  [Refereed]

  • Non-Asymptotic Fundamental Limits of Guessing Subject to Distortion

    Shota Saito, Toshiyasu Matsushima

    2019 IEEE International Symposium on Information Theory    2019.07  [Refereed]

  • Distributed Stochastic Gradient Descent Using LDGM Codes

    堀井俊佑, 吉田隆弘, 小林学, 松嶋敏泰

    Proceedings of 2019 IEEE International Symposium on Information Theory (ISIT2019)     1417 - 1421  2019.07  [Refereed]

    DOI

  • 推薦システムの新規顧客問題における半教師付き学習

    前田康成, 山内翔, 鈴木正清, 松嶋敏泰

    バイオメディカル・ファジィ・システム学会誌   20 ( 1 ) 37 - 46  2018

  • 顧客クラスが変化する推薦システムにおける半教師付き学習

    前田康成, 山内翔, 鈴木正清, 松嶋敏泰

    バイオメディカル・ファジィ・システム学会誌   20 ( 1 ) 15 - 22  2018

  • 推薦システムにおける新規顧客問題に関する一考察

    前田康成, 山内翔, 鈴木正清, 松嶋敏泰

    バイオメディカル・ファジィ・システム学会誌   Vol.19 ( 2 ) 21 - 27  2017.12  [Refereed]

  • マルコフ決定過程を用いたヘルスケア支援に関する一考察

    前田康成, 山内翔, 鈴木正清, 高野賢裕, 松嶋敏泰

    バイオメディカル・ファジィ・システム学会誌   Vol.19 ( 2 ) 21 - 27  2017.12  [Refereed]

  • Collaborative Filtering Based on the Latent Class Model for Attributes

    Manabu Kobayashi, Kenta Mikawa, Masayuki Goto, Toshiyasu Matsushima, Shigeichi Hirasawa

      ( 893 ) 896 - 201712  2017.12  [Refereed]

  • Spatially "Mt. Fuji" Coupled LDPC codes

    Yuta Nakahara, Shota Saito, Toshiyasu Matsushima

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E100A ( 12 ) 2594 - 2606  2017.12

     View Summary

    A new type of spatially coupled low density parity check (SCLDPC) code is proposed. This code has two benefits. (1) This code requires less number of iterations to correct the erasures occurring through the binary erasure channel in the waterfall region than that of the usual SCLDPC code. (2) This code has lower error floor than that of the usual SCLDPC code. Proposed code is constructed as a coupled chain of the underlying LDPC codes whose code lengths exponentially increase as the position where the codes exist is close to the middle of the chain. We call our code spatially "Mt. Fuji" coupled LDPC (SFCLDPC) code because the shape of the graph representing the code lengths of underlying LDPC codes at each position looks like Mt. Fuji. By this structure, when the proposed SFCLDPC code and the original SCLDPC code are constructed with the same code rate and the same code length, L (the number of the underlying LDPC codes) of the proposed SFCLDPC code becomes smaller and M (the code lengths of the underlying LDPC codes) of the proposed SFCLDPC code becomes larger than those of the SCLDPC code. These properties of L and M enables the above reduction of the number of iterations and the bit error rate in the error floor region, which are confirmed by the density evolution and computer simulations.

    DOI

  • Evaluation of overflow probability of bayes code in moderate deviation regime

    Shota Saito, Toshiyasu Matsushima

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E100A ( 12 ) 2728 - 2731  2017.12

     View Summary

    This letter treats the problem of lossless fixed-to-variable length source coding in moderate deviation regime. We investigate the behavior of the overflow probability of the Bayes code. Our result clarifies that the behavior of the overflow probability of the Bayes code is similar to that of the optimal non-universal code for i.i.d. sources.

    DOI

  • A Study on Analytical Properties of Bayesian Experimental Design Model based on an Orthonormal System

    Yoshifumi Ukita, Shunsuke Horii, Toshiyasu Matsushima

    Proceedings of Bayse on the Beach 2017     22 - 22  2017.11  [Refereed]

  • 復元および再生成の条件を一般化した再生成符号とその構成法

    鎌塚明, 東優太, 吉田隆弘, 松嶋敏泰

    電子情報学会論文誌 A   Vol.J100-A ( 11 ) 411 - 420  2017.11  [Refereed]

  • 2-レベル不均一誤り訂正符号の線形計画限界

    斎藤友彦, 新家稔央, 浮田善文, 松嶋敏泰, 平澤茂一

    電子情報学会論文誌 A   Vol.J100-A ( 9 ) 316 - 324  2017.09  [Refereed]

  • Variable-length lossy compression allowing positive overflow and excess distortion probabilities

    Shota Saito, Hideki Yagi, Toshiyasu Matsushima

    IEEE International Symposium on Information Theory - Proceedings     1568 - 1572  2017.08

     View Summary

    This paper investigates the problem of variable-length lossy source coding. We deal with the case where both the excess distortion probability and the overflow probability of codeword length are less than or equal to positive constants. The infimum of the thresholds on the overflow probability is characterized by a smooth max entropy-based quantity. Both non-asymptotic and asymptotic cases are analyzed. To show the achievability results, we do not utilize the random coding argument but give an explicit code construction.

    DOI

  • Evaluation of overflow probability of Bayes code in moderate deviation regime

    Saito, Shota, Matsushima, Toshiyasu

    Proceedings of 2016 International Symposium on Information Theory and Its Applications, ISITA 2016     1 - 5  2017.02

  • Threshold of overflow probability in terms of smooth max-entropy for variable-length compression allowing errors

    Saito, Shota, Matsushima, Toshiyasu

    Proceedings of 2016 International Symposium on Information Theory and Its Applications, ISITA 2016     21 - 25  2017.02

  • Spatially 'Mt. Fuji' coupled LDPC codes

    Nakahara, Yuta, Saito, Shota, Matsushima, Toshiyasu

    Proceedings of 2016 International Symposium on Information Theory and Its Applications, ISITA 2016     201 - 205  2017.02

  • A note on asset management with sensor network

    Yasunari Maeda, Masakiyo Suzuki, Toshiyasu Matsushima

    IEEJ Transactions on Electronics, Information and Systems   137 ( 6 ) 817 - 818  2017

     View Summary

    In this research we apply Markov decision processes to asset management with sensor network. We propose a new asset management method which minimizes total cost with reference to a Bayes criterion.

    DOI CiNii

  • A note on recommender system with transitions of user classes

    Yasunari Maeda, Masakiyo Suzuki, Toshiyasu Matsushima

    IEEJ Transactions on Electronics, Information and Systems   137 ( 6 ) 815 - 816  2017

     View Summary

    In this research we apply Markov decision processes to recommender system with transitions of user classes. We propose a new recommender method which maximizes total reward with reference to a Bayes criterion.

    DOI CiNii

  • 半教師付き学習における一致性を満たすゆう度方程式の解に基づく予測の漸近評価

    安田豪毅, 宮希望, 須子統太, 松嶋敏泰

    電子情報通信学会論文誌A   Vol.J100-A ( 1 ) 102 - 113  2017.01

  • A note on asset management with sensor network

    Yasunari Maeda, Masakiyo Suzuki, Toshiyasu Matsushima

    IEEJ Transactions on Electronics, Information and Systems   137 ( 6 ) 817 - 818  2017  [Refereed]

     View Summary

    In this research we apply Markov decision processes to asset management with sensor network. We propose a new asset management method which minimizes total cost with reference to a Bayes criterion.

    DOI

  • A note on recommender system with transitions of user classes

    Yasunari Maeda, Masakiyo Suzuki, Toshiyasu Matsushima

    IEEJ Transactions on Electronics, Information and Systems   137 ( 6 ) 815 - 816  2017  [Refereed]

     View Summary

    In this research we apply Markov decision processes to recommender system with transitions of user classes. We propose a new recommender method which maximizes total reward with reference to a Bayes criterion.

    DOI

  • Threshold of Overflow Probability Using Smooth Max-Entropy in Lossless Fixed-to-Variable Length Source Coding for General Sources

    Shota Saito, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E99A ( 12 ) 2286 - 2290  2016.12  [Refereed]

     View Summary

    We treat lossless fixed-to-variable length source coding under general sources for finite block length setting. We evaluate the threshold of the overflow probability for prefix and non-prefix codes in terms of the smooth max-entropy. We clarify the difference of the thresholds between prefix and non-prefix codes for finite block length. Further, we discuss our results under the asymptotic block length setting.

    DOI CiNii

  • Second-Order Achievable Rate Region of Slepian-Wolf Coding Problem in terms of Smooth Max-Entropy for General Sources

    Shota Saito, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E99A ( 12 ) 2275 - 2280  2016.12  [Refereed]

     View Summary

    This letter deals with the Slepian-Wolf coding problem for general sources. The second-order achievable rate region is derived using quantity which is related to the smooth max-entropy and the conditional smooth max-entropy. Moreover, we show the relationship of the functions which characterize the second-order achievable rate region in our study and previous study.

    DOI CiNii

  • Linear Programming Decoding of Binary Linear Codes for Symbol-Pair Read Channel

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E99A ( 12 ) 2170 - 2178  2016.12  [Refereed]

     View Summary

    In this study, we develop a new algorithm for decoding binary linear codes for symbol-pair read channels. The symbol-pair read channel was recently introduced by Cassuto and Blaum to model channels with higher write resolutions than read resolutions. The proposed decoding algorithm is based on linear programming (LP). For LDPC codes, the proposed algorithm runs in time polynomial in the codeword length. It is proved that the proposed LP decoder has the maximum-likelihood (ML) certificate property, i.e., the output of the decoder is guaranteed to be the ML codeword when it is integral. We also introduce the fractional pair distance d(fp) of the code, which is a lower bound on the minimum pair distance. It is proved that the proposed LP decoder corrects up to. inverted left perpendiculard(fp)/2inverted right perpendicular - 1 errors.

    DOI CiNii

  • Threshold of overflow probability using smooth max-entropy in lossless fixed-to-variable length source coding for general sources

    Shota Saito, Toshiyasu Matsushima

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E99A ( 12 ) 2286 - 2290  2016.12

     View Summary

    We treat lossless fixed-to-variable length source coding under general sources for finite block length setting. We evaluate the threshold of the overflow probability for prefix and non-prefix codes in terms of the smooth max-entropy. We clarify the difference of the thresholds between prefix and non-prefix codes for finite block length. Further, we discuss our results under the asymptotic block length setting.

    DOI

  • Second-order achievable rate region of Slepian-Wolf coding problem in terms of smooth max-entropy for general sources

    Saito, Shota, Matsushima, Toshiyasu

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E99A ( 12 ) 2275 - 2280  2016.12

     View Summary

    Copyright © 2016 The Institute of Electronics, Information and Communication Engineers.This letter deals with the Slepian-Wolf coding problem for general sources. The second-order achievable rate region is derived using quantity which is related to the smooth max-entropy and the conditional smooth max-entropy. Moreover, we show the relationship of the functions which characterize the second-order achievable rate region in our study and previous study.

    DOI

  • Linear programming decoding of binary linear codes for symbol-pair read channel

    Horii, Shunsuke, Matsushima, Toshiyasu, Hirasawa, Shigeichi

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E99A ( 12 ) 2170 - 2178  2016.12

     View Summary

    Copyright © 2016 The Institute of Electronics, Information and Communication Engineers.In this study, we develop a new algorithm for decoding binary linear codes for symbol-pair read channels. The symbol-pair read channel was recently introduced by Cassuto and Blaum to model channels with higher write resolutions than read resolutions. The proposed decoding algorithm is based on linear programming (LP). For LDPC codes, the proposed algorithm runs in time polynomial in the codeword length. It is proved that the proposed LP decoder has the maximum-likelihood (ML) certificate property, i.e., the output of the decoder is guaranteed to be the ML codeword when it is integral. We also introduce the fractional pair distance df p of the code, which is a lower bound on the minimum pair distance. It is proved that the proposed LP decoder corrects up to df p =2 - 1 errors.

    DOI

  • Linear Programming Decoding of Binary Linear Codes for Symbol-Pair Read Channels

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    2016 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY     1944 - 1948  2016  [Refereed]

     View Summary

    In this paper, we develop a new decoding algorithm of binary linear codes for symbol-pair read channel. The Symbol-pair read channel has recently been introduced by Cassuto and Blaum to model channel whose write resolution is higher than read resolution. The proposed decoding algorithm is based on the linear programming (LP). It is proved that the proposed LP decoder has the maximum-likelihood (ML) certificate property, i.e., the output of the decoder is guaranteed to be the ML codeword when it is integral. We also introduce the fractional pair distance d(fp) of the code which is a lower bound on the minimum pair distance. It is proved that the proposed LP decoder corrects up to [d(fp)/2] - 1 errors.

  • Theoretical Limit of Type-I Hybrid Selective-repeat ARQ with Finite Receiver Buffer

    Yasunari Maeda, Toshiyasu Matsushima

    PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016)     582 - 585  2016  [Refereed]

     View Summary

    There are many research on hybrid selective-repeat ARQ method. In many previous research purposes have been to maximize a throughput on the number of messages delivered to the user. Important theoretical limits were studied under the condition that a decoding error was very small and neglected.
    In this research we propose the optimal type-I hybrid selective-repeat ARQ with finite receiver buffer. The proposed method maximizes an expected total reward regarding the number of messages delivered to the user correctly based on statistical decision theory.

  • A Note on Support Recovery of Sparse Signals using Linear Programming

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016)     270 - 274  2016  [Refereed]

     View Summary

    A new theory known as compressed sensing considers the problem to acquire and recover a sparse signal from its linear measurements. In this paper, we propose a new support recovery algorithm from noisy measurements based on the linear programming (LP). LP is widely used to estimate sparse signals, however, we focus on the problem to recover the support of sparse signals rather than the problem to estimate sparse signals themselves. First, we derive an integer linear programming (ILP) formulation for the support recovery problem. Then we obtain the LP based support recovery algorithm by relaxing the ILP. The proposed LP based recovery algorithm has an attracting property that the output of the algorithm is guaranteed to be the maximum a posteiori (MAP) estimate when it is integer valued. We compare the performance of the proposed algorithm to a state-of-the-art algorithm named sparse matching pursuit (SMP) via numerical simulations.

  • A Maximum Likelihood Decoding Algorithm of Gabidulin Codes in Deterministic Network Coding

    Koki Kazama, Akira Kamatsuka, Toshiyasu Matsushima

    PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016)     666 - 670  2016  [Refereed]

     View Summary

    Recently, studies of linear network coding have attracted attention. This paper considers a probabilistic error model for the deterministic linear network coding. In the previous studies of error-correcting codes, Kaneko et al. proposed a decoding algorithms for probabilistic error model. This is a maximum likelihood decoding algorithm that uses the optimization method called branch and bound method. This paper constructs a new model of deterministic linear network coding and proposes a maximum likelihood decoding algorithm that uses the branch and bound method.

  • Relationships between Correlation of Information Stored on Nodes and Coding Efficiency for Cooperative Regenerating Codes

    Takahiro Yoshida, Toshiyasu Matsushima

    PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016)     31 - 35  2016  [Refereed]

     View Summary

    Cooperative regenerating codes are a class of codes that enable a data collector to reconstruct the original data by connecting to a subset of storage nodes, and also can repair multiple failed nodes by downloading data from the surviving nodes and exchanging data among the new nodes. In cooperative regenerating codes, there exists a tradeoff between the storage size of each node and repair-bandwidth. In this study, we propose new classes of cooperative regenerating codes based on correlation of information stored on nodes. We also consider relationships between each class of cooperative regenerating codes and storage size and repair-bandwidth.

  • A Note on Unequal Elinor Protection in Random Network Coding

    Tomohiko Saito, Koki Kazama, Toshihiro Niinomi, Toshiyasu Matsushima

    PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016)     661 - 665  2016  [Refereed]

     View Summary

    Linear Unequal Error Protection (UEP) codes were proposed by Masnick et al. and largely developed by Dunning et al. and Gils. On the other hand, random linear network coding is recently studied by many researchers. Kotter et al. proposed a new coding technique named subspace codes and applied it to random linear network coding. Moreover. Silva et al. proposed constructions of subspace codes using rank metric codes. In this paper, we propose subspace UEP codes and apply them to the random linear network coding. Then, we propose rank metric UEP codes and constructions of the subspace UEP codes using the rank metric UEP codes.

  • Regenerating Codes with Generalized Conditions of Reconstruction and Regeneration

    Akira Kamatsuka, Yuta Azuma, Takahiro Yoshida, Toshiyasu Matsushima

    PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016)     41 - 45  2016  [Refereed]

     View Summary

    As a distributed storage system with regenerating function of failed nodes, regenerating codes have been studied recently. In a [n, k, d]-regenerating codes framework, a message can be reconstructed from any subsets of k nodes out of n nodes and any failed nodes can be repaired from any subsets of d nodes out of (n -1) nodes. The conditions of the reconstruction and regeneration can be generalized. In this paper, we propose the regenerating codes with generalized conditions and derive their optimal constructing algorithm under the multiple assignment map construction framework. The optimal codes of the framework are obtained by solving integer programming problems in a similar way of generalized secret sharing scheme's construction based on a multiple assignment map.

  • Evaluation of the Bayes Code from Viewpoints of the Distribution of Its Codeword Lengths

    Shota Saito, Nozomi Miya, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E98A ( 12 ) 2407 - 2414  2015.12  [Refereed]

     View Summary

    This paper considers universal lossless variable-length source coding problem and investigates the Bayes code from viewpoints of the distribution of its codeword lengths. First, we show that the codeword lengths of the Bayes code satisfy the asymptotic normality. This study can be seen as the investigation on the asymptotic shape of the distribution of codeword lengths. Second, we show that the codeword lengths of the Bayes code satisfy the law of the iterated logarithm. This study can be seen as the investigation on the asymptotic end points of the distribution of codeword lengths. Moreover, the overflow probability, which represents the bottom of the distribution of codeword lengths, is studied for the Bayes code. We derive upper and lower bounds of the infimum of a threshold on the overflow probability under the condition that the overflow probability does not exceed epsilon is an element of(0, 1). We also analyze the necessary and sufficient condition on a threshold for the overflow probability of the Bayes code to approach zero asymptotically.

    DOI

  • On the Credible Interval Prediction of Web Traffic based on the Bayes Decision Theory

    Daiki Koizumi, Toshiyasu Matsushima

    Proceeding of the 38th Symposium on Information Theory and its Applications (SITA2015)     235 - 240  2015.11

  • On the Bayes Optimal Prediction of Time Series under a Nonstationary Weibull Distribution

    Daiki Koizumi, Toshiyasu Matsushima

    IEICE Technical Report (IT2015-33)   115 ( 137 ) 95 - 100  2015.07

    CiNii

  • Fundamental Limit and Pointwise Asymptotics of the Bayes Code for Markov Sources

    Shota Saito, Nozomi Miya, Toshiyasu Matsushima

    2015 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT)     1986 - 1990  2015

     View Summary

    This paper considers universal lossless variablelength source coding problem and deals with one of the fundamental limits and pointwise asymptotics of the Bayes code for stationary ergodic finite order Markov sources. As investigation of the fundamental limits, we show upper and lower bounds of the minimum rate such that the probability which exceeds it is less than epsilon is an element of (0, 1). Furthermore, we prove that the codeword length of the Bayes code satisfies the asymptotic normality (pointwise root n asymptotics) and the law of the iterated logarithm (pointwise root n log log n asymptotics), where n represents length of a source sequence and "log" is the natural logarithm.

  • Evaluation of the Bayes Code from Viewpoints of the Distribution of Its Codeword Lengths

    SAITO Shota, MIYA Nozomi, MATSUSHIMA Toshiyasu

    IEICE Trans. Fundamentals   98 ( 12 ) 2407 - 2414  2015

     View Summary

    This paper considers universal lossless variable-length source coding problem and investigates the Bayes code from viewpoints of the distribution of its codeword lengths. First, we show that the codeword lengths of the Bayes code satisfy the asymptotic normality. This study can be seen as the investigation on the asymptotic shape of the distribution of codeword lengths. Second, we show that the codeword lengths of the Bayes code satisfy the law of the iterated logarithm. This study can be seen as the investigation on the asymptotic end points of the distribution of codeword lengths. Moreover, the overflow probability, which represents the bottom of the distribution of codeword lengths, is studied for the Bayes code. We derive upper and lower bounds of the infimum of a threshold on the overflow probability under the condition that the overflow probability does not exceed ε∈(0,1). We also analyze the necessary and sufficient condition on a threshold for the overflow probability of the Bayes code to approach zero asymptotically.

    CiNii

  • Welcome message from the IECON2015 general chairs

    Kiyoshi Ohishi, Hideki Hashimoto, Toshiyuki Murakami, Leopoldo G. Franquelo, Mo-Yuen Chow

    IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society     1 - 2  2015

    DOI

  • Foreword: Special section on information theory and its applications

    Toshiyasu Matsushima, Manabu Kobayashi, Shiro Ikeda, Shogo Usami, Kenta Kasai, Shigeaki Kuzuoka, Hiroki Koga, Tetsuya Kojima, Masayuki Goto, Tatsumi Konishi, Hidetoshi Saito, Ryuichi Sakai, Mikihiko Nishiara, Ryo Nomura, Mitsuru Hamada, Masaya Fujisawa, Tetsunao Matsuta, Ryutaroh Matsumoto, Kazuhiko Minematsu, Kazushi Mimura, Hideki Yagi

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E97A ( 12 ) 2287  2014.12  [Refereed]

  • Asymptotics of Bayesian Inference for a Class of Probabilistic Models under Misspecification

    Nozomi Miya, Tota Suko, Goki Yasuda, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E97A ( 12 ) 2352 - 2360  2014.12  [Refereed]

     View Summary

    In this paper, sequential prediction is studied. The typical assumptions about the probabilistic model in sequential prediction are following two cases. One is the case that a certain probabilistic model is given and the parameters are unknown. The other is the case that not a certain probabilistic model but a class of probabilistic models is given and the parameters are unknown. If there exist some parameters and some models such that the distributions that are identified by them equal the source distribution, an assumed model or a class of models can represent the source distribution. This case is called that specifiable condition is satisfied. In this study, the decision based on the Bayesian principle is made for a class of probabilistic models (not for a certain probabilistic model). The case that specifiable condition is not satisfied is studied. Then, the asymptotic behaviors of the cumulative logarithmic loss for individual sequence in the sense of almost sure convergence and the expected loss, i.e. redundancy are analyzed and the constant terms of the asymptotic equations are identified.

    DOI

  • A Note on Improvement in the Rate of a Prediction Error of AdaBoost in Pattern Recognition

    Hideyuki Masui, Ryoma Tsuzuki, Nozomi Miya, Toshiyasu Matsushima

    IPSJ SIG Notes. CVIM   2014 ( 1 ) 1 - 6  2014.08

     View Summary

    AdaBoost is an algorithm used for pattern recognition. This algorithm successively makes the model which minimizes an error for sample data. However, to consider prediction, there is no guarantee that such model is optimum. Regarding this issue, there are many studies about regularization method and variable selection method. Forward selection method is one of the methods of variable selection method and Stagewise is one of the methods of forward selection method. Each iteration, Stagewise selects one variable. Stagewise updates parameters slightly increase and not minimizes an error for sample data. It is known that Stagewise improves the rate of a prediction error. In this research, we apply the idea of Stagewise for AdaBoost and consider improvement the rate of a prediction error.

    CiNii

  • A Note on the Correlated Multiple Matrix Completion based on the Convex Optimization Method

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    2014 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC)     1618 - 1623  2014  [Refereed]

     View Summary

    In this paper, we consider a completion problem of multiple related matrices. Matrix completion problem is the problem to estimate unobserved elements of the matrix from observed elements. It has many applications such as collaborative filtering, computer vision, biology, and so on. In cases where we can obtain some related matrices, we can expect that their simultaneous completion has better performance than completing each matrix independently. Collective matrix factorization is a powerful approach to jointly factorize multiple matrices. However, existing completion algorithms for the collective matrix factorization have some drawbacks. One is that most existing algorithms are based on non-convex formulations of the problem. Another is that only a few existing algorithms consider the strength of the relation among matrices and it results in worse performance when some matrices are actually not related. In this paper, we formulate the multiple matrix completion problem as the convex optimization problem. Moreover, it considers the strength of the relation among matrices. We also develop an optimization algorithm which solves the proposed problem efficiently based on the alternating direction method of multipliers (ADMM). We verify the effectiveness of our approach through numerical experiments on both synthetic data and real data set: MovieLens.

  • Evaluation of the Minimum Overflow Threshold of Bayes Codes for a Markov Source

    Shota Saito, Nozomi Miya, Toshiyasu Matsushima

    2014 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA)     211 - 215  2014

     View Summary

    The objective of this research is to evaluate the epsilon-minimum overflow threshold of the Bayes codes for a Markov source. In the lossless variable-length source coding problem, typical criteria are the mean codeword length and the overflow probability. The overflow probability is the probability with which a codeword length per source symbol exceeds a threshold and the epsilon-minimum overflow threshold is defined. In the non-universal setting, the Shannon code is optimal under the mean codeword length and the epsilon-minimum overflow threshold for the Shannon code is derived for an i.i.d. source. On the other hand, in the universal setting, the Bayes code is one of universal codes which minimize the mean codeword length under the Bayes criterion. However, few studies have been done on the overflow probability for the Bayes codes. In this paper, we assume a stationary ergodic finite order Markov source and derive the upper and lower bounds of the epsilon-minimum overflow threshold of the Bayes codes.

  • Privacy-preserving Distributed Calculation Methods of a Least-squares Estimator for Linear Regression Models

    SUKO Tota, HORII Shunsuke, KOBAYASHI Manabu, GOTO Masayuki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    J. Jpn . Ind. Manage. Assoc.   65 ( 2 ) 78 - 88  2014

     View Summary

    In this paper, we study a privacy preserving linear regression analysis. We propose a new protocol of a distributed calculation method that calculates a least squares estimator, in the case that two parties have different types of explanatory variables. We show the security of privacy in the proposed protocol. Because the protocol have iterative calculations, we evaluate the number of iterations via numerical experiments. Finally, we show an extended protocol that is a distributed calculation method for k parties.

    CiNii

  • The Empirical Bayes Forecasting of Web Traffic on the Non-Stationary Poisson Distribution

    Daiki Koizumi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proceeding of the 36th Symposium on Information Theory and its Applications (SITA2013)     432 - 436  2013.11

  • A Note on A Strategy for Role-Playing Game with Nonplayer Characters

    MAEDA Yasunari, GOTO Fumitaro, MASUI Hiroshi, MASUI Fumito, SUZUKI Masakiyo, MATSUSHIMA Toshiyasu

    The Transactions of the Institute of Electronics, Information and Communication Engineers. A   96 ( 8 ) 572 - 581  2013.08

    CiNii

  • Active Learning Strategy for Role-playing Game Modeled by Markov Decision Processes

    MAEDA Yasunari, GOTO Fumitaro, MASUI Hiroshi, MASUI Fumito, SUZUKI Masakiyo, MATSUSHIMA Toshiyasu

    Journal of Biomedical Fuzzy Systems Association   15 ( 1 ) 69 - 81  2013.06

     View Summary

    In previous research a role-playing game(RPG) is represented with Markov decision processes(MDP). But active learning method for RPG has not been studied yet. In this research we propose an active learning method which maximizes an expected total reward with respect to a Bayes criterion under the condition that the true parameter of MDP is unknown. We recognize the effectiveness of our proposed method by some simulations.

    CiNii

  • Iterative multiuser joint decoding based on ADMM

    S. Horii, T. Suko, T. Matsushima, S. Hirasawa

    2013 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2013 - Proceedings     1097 - 1100  2013

     View Summary

    In this paper, we develop an iterative multiuser joint decoding of code-division multiple-access (CDMA) signals based on a distributed optimization algorithm. For the joint decoding problem, decoding algorithm based on the turbo principle is widely used. The algorithm consists of soft-input soft-output (SISO) channel decoder and SISO multiuser detector and it can be derived as an application of the sum-product algorithm. On the other hand, in the research area of error correcting codes, the decoding algorithm based on convex optimization has been attracting a great deal of attention. Decoding algorithm based on linear programming (LP) has decoding error rate which is comparable with sum-product algorithm with stronger theoretical guarantees. We formulate the joint decoding problem of CDMA signals as an convex optimization problem and we present a relax form of the problem. Moreover, we propose a distributed algorithm which efficiently solves the relaxed optimization problem. The proposed algorithm is based on alternating direction method of multipliers (ADMM). We also see the performance of the proposed decoder through numerical simulations. © 2013 IEEE.

    DOI

  • An Optimal Prediction Method Using Hierarchical N-gram Based on Bayesian Decision Theory

    Takashi Suenaga, Toshiyasu Matsushima

      vol.6 ( no.1 ) 102 - 110  2013  [Refereed]

     View Summary

    Predictive word is an input technology showing candidate words which a system predict by user partial input. We treat predictive methods using an N-gram model. The model is generally produced by analyzing train data. The data is more sparse in proportion to an N-gram order, because of enormous combinations of words in the sequences. An issue of producing the model is how to combine a lower order model into a higher order one. Many researchers proposed models composed of weighed each-order one, such as a mixture distribution or an interpolation created by discount parameters considering about extremely lower frequent sequence. But these methods have no theoretical guarantee about prediction errors. In this paper, we treat the issue as a statistical problem that the model order is unknown, and discuss prediction errors from a point of view about Bayesian decision theory. We present that an optimal prediction method with reference to the Bayes criterion for minimizing the errors. Experimental results using Japanese documents show that our method performs good predictive words.

  • A study on the degrees of freedom in an experimental design model based on an orthonormal system

    Yoshifumi Ukita, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E96-A ( 2 ) 658 - 662  2013

     View Summary

    Experiments usually aim to study how changes in various factors affect the response variable of interest. Since the response model used most often at present in experimental design is expressed through the effect of each factor, it is straightforward to ascertain how each factor affects the response variable. However, since the response model contains redundant parameters, in the analysis of variance we must calculate the degrees of freedom defined by the number of independent parameters. In this letter, we propose the idea of calculating the degrees of freedom over the model based on an orthonormal system for the first time. In this way, we can easily obtain the number of independent parameters associated with any component, which reduces the risk of mistakes in the calculation of the number of independent parameters and facilitates the implementation of estimation procedures. Copyright © 2013 The Institute of Electronics, Information and Communication Engineers.

    DOI

  • A Generalized Ramp Scheme for Key Distribution Scheme and an Optimal Construction

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    The Transactions of the Institute of Electronics, Information and Communication Engineers. A   95 ( 10 ) 723 - 736  2012.10

    CiNii

  • An Error Probability Estimation of the Document Classification Using Markov Model

    Manabu Kobayashi, Hiroshi Ninomiya, Toshiyasu Matsushima, Shigeichi Hirasawa

    2012 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2012)     717 - 721  2012  [Refereed]

     View Summary

    The document classification problem has been investigated by various techniques, such as a vector space model, a support vector machine, a random forest, and so on. On the other hand, J.Ziv et al. have proposed a document classification method using Ziv-Lempel algorithm to compress the data. Furthermore, the Context-Tree Weighting (CTW) algorithm has been proposed as an outstanding data compression, and for the document classification using the CTW algorithm experimental results have been reported. In this paper, we assume that each document with same category arises from Markov model with same parameters for the document classification. Then we propose an analysis method to estimate a classification error probability for the document with the finite length.

  • Fault Diagnosis Algorithm in Multi-Computer Systems based on Lagrangian Relaxation Method

    Shunsuke Horii, Manabu Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    2012 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2012)     712 - 716  2012

     View Summary

    We propose new algorithms for fault diagnosis problem based on the dual decomposition method and the augmented Lagrangian method. Our algorithms are convergent and those outputs are same as that of Linear Programming (LP) based fault diagnosis algorithm. The proposed algorithms have smaller computational complexity than ordinary LP solver. Experimental results show the practical potentials of the proposed algorithms.

  • Information Spectrum Approach to Overflow Probability of Variable-Length Codes with Conditional Cost Function

    Ryo Nomura, Toshiyasu Matsushima

    2012 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS (ISIT)    2012

     View Summary

    Lossless variable-length source coding with unequal cost function is considered for general sources. In this problem, the codeword cost instead of codeword length is important. The infimum of average codeword cost has already been determined for general sources. We consider the overflow probability of codeword cost and determine the infimum of achievable overflow threshold. Our analysis is on the basis of information-spectrum methods and hence valid through the general source.

  • Asymptotics of Bayesian estimation for nested models under misspecification

    Nozomi Miya, Tota Suko, Goki Yasuda, Toshiyasu Matsushima

    2012 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2012)     86 - 90  2012

     View Summary

    We analyze the asymptotic properties of the cumulative logarithmic loss in the decision problem based on the Bayesian principle and explicitly identify the constant terms of the asymptotic equations as in the case of previous studies by Clarke and Barron and Gotoh et al. We assume that the set of models is given that identify a class of parameterized distributions, it has a nested structure and the source distribution is not contained in all the families of parameterized distributions that are identified by each model. The cumulative logarithmic loss is the sum of the logarithmic loss functions for each time decision-, e. g., the redundancy in the universal noiseless source coding.

  • The Optimal Key Estimation of Stream Ciphers and Its Approximation Algorithm Based on a Probabilistic Inference

    Yuji Iikubo, Shunsuke Horii, Toshiyasu Matsushima

    2012 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2012)     531 - 535  2012

     View Summary

    A stream cipher is an important class of encryption algorithms. Its safety depends on the structure of the pseudo-random number generator used. There are various types of pseudo-random number generators in existence, and attack algorithms used on them have been studied individually. In this paper, we express the problem of attacks on a general stream cipher as a probabilistic inference problem, and formulate the optimal key estimation. We also propose a unified framework of attack algorithms that can be applied to a wide variety of stream ciphers. The optimal key estimation, however, has computational complexity. To reduce the complexity, an approximation algorithm based on a probabilistic inference is proposed. We also describe some attack algorithms used on practical pseudo-random number generators. Finally, the proposed algorithm is evaluated by through a computer simulation.

  • Variable Order Transition Probability Markov Decision Process for the Recommendation System

    Shuhei Kuwata, Yasunari Maeda, Toshiyasu Matsushima, Shigeichi Hirasawa

      vol.6 ( no.1 ) 20 - 30  2012  [Refereed]

     View Summary

    In this paper, we propose a general markov decision process model for the recommendation system. Furthermore, by using historical data, we derive the optimal recommendation lists from the proposed model based on bayesian decision theory. In the recommendation research area, there were little studies that considered both the purchased items and the past recommended items within a given period. In these circumstances, markov decision process based recommend method that can take these two things into account has been proposed. Our method also uses both things as with the previous method. Here, the unique thing about this paper is not only that we generalize the existing model, but also that we formulate the process to get the recommendation lists as the statistical decision problem. As a result, we can obtain the most suitable recommendation lists with respect to the purpose of the recommendation for a wide variety of recommendation scene. By using artificial data, we show the experimental results that our method can obtain more rewards than the conventional method gets.

  • A Note on Relation between the Fourier Coefficients and the Effects in the Experimental Design

    Ukita, Yoshifumi, Matsushima, Toshiyasu

    Journal of Communication and Computer   vol.9 ( no.7 ) 830 - 836  2012  [Refereed]

     View Summary

    It has recently been shown that the model in experimental design can be expressed in terms of an orthonormal system. In this case, the model is expressed by using Fourier coefficients instead of the effect of each factor. As there is an abundance of software for calculating the Fourier transform, such a system allows for a straightforward implementation of the procedures for estimating the Fourier coefficients by using Fourier transform. However, Fourier coefficients themselves do not provide a direct representation of the effect of each factor, and the relation between the Fourier coefficients and the effect of each factor has not yet been clarified. In this paper, we present theorems of the relation between the Fourier coefficients and the effect of each factor. By using these theorems, the effect of each factor can be easily obtained from the computed Fourier coefficients. Therefore, with the aid of an orthonormal system, it is possible to easily implement the estimation procedures as well as to understand how each factor affects the response variable in the model.

  • Berkeley Report

    MATSUSHIMA Toshiyasu

    IEICE Fundamentals Review   5 ( 3 ) 281 - 282  2012

    CiNii

  • An Analysis of Slepian-Wolf Coding Problem Based on the Asymptotic Normality

    Ryo Nomura, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E94A ( 11 ) 2220 - 2225  2011.11  [Refereed]

     View Summary

    Source coding theorem reveals the minimum achievable code length under the condition that the error probability is smaller than or equal to some small constant. In the single user communication system, the source coding theorem was proved for general sources. The class of general source is quite large and it is important result since the result can be applied for a wide class of sources. On the other hand there are several studies to evaluate the achievable code length more precisely for the restricted class of sources by using the restriction. In the multi-user communication system, although the source coding theorem was proved for general correlated sources, there is no study to evaluate the achievable code length more precisely. In this study, we consider the stationary memoryless correlated sources and show the coding theorem for Slepian-Wolf type problem more precisely than the previous result.

    DOI

  • On the Overflow Probability of Fixed-to-Variable Length Codes with Side Information

    Ryo Nomura, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E94A ( 11 ) 2083 - 2091  2011.11  [Refereed]

     View Summary

    The overflow probability is one of criteria that evaluate the performance of fixed-to-variable length (FV) codes. In the single source coding problem, there were many researches on the overflow probability. Recently, the source coding problem for correlated sources, such as Slepian-Wolf coding problem or source coding problem with side information, is one of main topics in information theory. In this paper, we consider the source coding problem with side information. In particular, we consider the FV code in the case that the encoder and the decoder can see side information. In this case, several codes were proposed and their mean code lengths were analyzed. However, there was no research about the overflow probability. We shall show two lemmas about the overflow probability. Then we obtain the condition that there exists a FV code under the condition that the overflow probability is smaller than or equal to some constant.

    DOI

  • A Note on the Linear Programming Decoding of Binary Linear Codes for Multiple-Access Channel

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E94A ( 6 ) 1230 - 1237  2011.06  [Refereed]

     View Summary

    In this paper, we develop linear-programming (LP) decoding for multiple-access channels with binary linear codes. For single-user channels, LP decoding has attracted much attention in recent years as a good approximation to maximum-likelihood (ML) decoding. We demonstrate how the ML decoding problem for multiple-access channels with binary linear codes can be formulated as an LP problem. This is not straightforward, because the objective function of the problem is generally a non-linear function of the codeword symbols. We introduce auxiliary variables such that the objective function is a linear function of these variables. The ML decoding problem then reduces to the LP problem. As in the case for single-user channels, we formulate the relaxed LP problem to reduce the complexity for practical implementation, and as a result propose a decoder that has the desirable property known as the ML certificate property (i.e., if the decoder outputs an integer solution, the solution is guaranteed to be the ML codeword). Although the computational complexity of the proposed algorithm is exponential in the number of users, we can reduce this complexity for Gaussian multiple-access channels. Furthermore, we compare the performance of the proposed decoder with a decoder based on the sum-product algorithm.

    DOI

  • Disk Allocation Methods for Cartesian Product Files Using Unequal Error Protection Codes

    Tomohiko Saito, Hiroshige Inazumi, Toshiyasu Matsushima, Shigeichi Hirasawa

    2011 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC)     2437 - 2442  2011  [Refereed]

     View Summary

    Allocation methods for Cartesian product files on multiple disks using linear error-correcting codes are discussed. In this paper, we propose an allocation method using unequal error protection (UEP) codes. Codewords of an UEP code have some special bits which are protected against a greater number of errors than the other bits. We firstly assume a model that "*", which means "don't care", appears with different probability in each attribute of queries. In this case, the average access time can be calculated using the split distance distribution. Finally, we illustrate the average access time of the methods using UEP codes.

  • Probabilistic Fault Diagnosis and its Analysis in Multicomputer Systems

    Manabu Kobayashi, Toshinori Takabatake, Toshiyasu Matsushima, Shigeichi Hirasawa

    2011 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC)   E101-A ( 12 ) 1205 - 1211  2011

     View Summary

    F.P.Preparata et al. have proposed a fault diagnosis model to find all faulty units in the multicomputer system by using outcomes which each unit tests some other units. In this paper, for probabilistic diagnosis models, we show an efficient diagnosis algorithm to obtain a posteriori probability that each of units is faulty given the test outcomes. Furthermore, we propose a method to analyze the diagnostic error probability of this algorithm.

  • A Note on the Degrees of Freedom in an Experimental Design Model Based on an Orthonormal System

    Yoshifumi Ukita, Toshiyasu Matsushima, Shigeichi Hirasawa

    2011 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC)     2181 - 2185  2011

     View Summary

    Experiments usually aim to study how changes in various factors affect the response variable of interest. Since the response model used most often at present in experimental design is expressed through the effect of each factor, it is straightforward to ascertain how each factor affects the response variable. However, since the response model contains redundant parameters, we must calculate the degrees of freedom defined by the number of independent parameters in the analysis of variance. In this paper, we show that through a description of experimental design based on an orthonormal system, the response model can be expressed using only independent parameters. Hence, we do not have to calculate the degrees of freedom defined by the number of independent parameters.

  • System Evaluation of Disk Allocation Methods for Cartesian Product Files by using Error Correcting Codes

    Shigeichi Hirasawa, Tomohiko Saito, Hiroshige Inazumi, Toshiyasu Matsushima

    2011 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC)     2443 - 2448  2011

     View Summary

    We discuss disk allocation methods for Cartesian product files by introducing error correcting codes, and have clarified the performance of the methods by system evaluation models developed by using rate distortion theory. Let us assume q(n) Cartesian product files with n attributes and q actual values in each attribute, and store q(n) files into G(<= q(n)) disks. For a partial match access request, we represent new disk allocation methods which able to access the disks in parallel as much as possible, where the partial match access request includes an indefinite case (don't care: "*") in some attributes and the * requires to access the files with corresponding to the attribute for the all actual attribute values. In this paper, we propose to apply unequal error protection codes to the case where the probabilities of occurrence of the * in the attributes for a partial match access request are not the same. We show the disk allocation methods have desirable properties as n becomes large.

  • A Description of Experimental Design on the Basis of an Orthonormal System

    Yoshifumi Ukita, Toshiyasu Matsushima

    Applications of Digital Signal Processing, Edited by Christian Cuadrado-Laborde, InTech Publisher     365 - 378  2011

     View Summary

    In this chapter, we propose that the model of experimental design be expressed as an orthonormal system, and show that the model contains no redundant parameters. Then, the model is expressed by using Fourier coefficients instead of the effect of each factor. As there is an abundance of software for calculating the Fourier transform, such a system allows for a straightforward implementation of the procedures for estimating the Fourier coefficients by using Fourier transform. In addition, the effect of each factor can be easily obtained from the Fourier coefficients. Therefore, it is possible to implement easily the estimation procedures as well as to understand how each factor affects the response variable in a model based on an orthonormal system. Moreover, the analysis of variance can also be performed in a model based on an orthonormal system. Hence, it is clear that two main procedures in the experimental design, that is, the estimation of the effects and the analysis of variance can be executed in a description of experimental design on the basis of an orthonormal system.

    DOI

  • 私たちの仕事と家庭のバランス

    松嶋 敏泰, 松嶋 智子

    電子情報通信学会 通信ソサイエティマガジン   2011 ( 16 ) 16_16 - 16_17  2011

    CiNii

  • A note on the linear programming decoding of binary linear codes for multiple-access channel

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E94-A ( 6 ) 1230 - 1237  2011

     View Summary

    In this paper, we develop linear-programming (LP) decoding for multiple-access channels with binary linear codes. For single-user channels, LP decoding has attracted much attention in recent years as a good approximation to maximum-likelihood (ML) decoding. We demonstrate how the ML decoding problem for multiple-access channels with binary linear codes can be formulated as an LP problem. This is not straightforward, because the objective function of the problem is generally a nonlinear function of the codeword symbols. We introduce auxiliary variables such that the objective function is a linear function of these variables. The ML decoding problem then reduces to the LP problem. As in the case for single-user channels, we formulate the relaxed LP problem to reduce the complexity for practical implementation, and as a result propose a decoder that has the desirable property known as the ML certificate property (i.e., if the decoder outputs an integer solution, the solution is guaranteed to be the ML codeword). Although the computational complexity of the proposed algorithm is exponential in the number of users, we can reduce this complexity for Gaussian multiple-access channels. Furthermore, we compare the performance of the proposed decoder with a decoder based on the sum-product algorithm. Copyright © 2011 The Institute of Electronics, Information and Communication Engineers.

    DOI

  • A note on document classification with small training data

    Yasunari Maeda, Hideki Yoshida, Masakiyo Suzuki, Toshiyasu Matsushima

    IEEJ Transactions on Electronics, Information and Systems   131 ( 8 ) 1459 - 1466  2011

     View Summary

    Document classification is one of important topics in the field of NLP (Natural Language Processing). In the previous research a document classification method has been proposed which minimizes an error rate with reference to a Bayes criterion. But when the number of documents in training data is small, the accuracy of the previous method is low. So in this research we use estimating data in order to estimate prior distributions. When the training data is small the accuracy using estimating data is higher than the accuracy of the previous method. But when the training data is big the accuracy using estimating data is lower than the accuracy of the previous method. So in this research we also propose another technique whose accuracy is higher than the accuracy of the previous method when the training data is small, and is almost the same as the accuracy of the previous method when the training data is big. © 2011 The Institute of Electrical Engineers of Japan.

    DOI CiNii

  • A Note on the Branch-and-Cut Approach to Decoding Linear Block Codes

    Shunsuke Horii, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E93A ( 11 ) 1912 - 1917  2010.11  [Refereed]

     View Summary

    Maximum likelihood (ML) decoding of linear block codes can be considered as an integer linear programming (ILP) Since it is an NP hard problem in general there are many researches about the algorithms to approximately solve the problem One of the most popular algorithms is linear programming (LP) decoding proposed by Feldman et al LP decoding is based on the LP relaxation which Is a method to approximately solve the ILP corresponding to the ML decoding problem Advanced algorithms for solving ILP (approximately or exactly) include cutting plane method and branch and bound method As applications of these methods adaptive LP decoding and branch and bound decoding have been proposed by Taghavi et al and Yang et al respectively Another method for solving ILP is the branch and cut method which is a hybrid of cutting plane and branch and bound methods The branch and cut method is widely used to solve ILP however it is unobvious that the method works well for the ML decoding problem In this paper we show that the branch and cut method is certainly effective for the ML decoding problem Furthermore the branch and cut method consists of some technical components and the performance of the algorithm depends on the selection of these components It is important to consider how to select the technical components in the branch and cut method We see the differences caused by the selection of those technical components and consider which scheme is most effective for the ML decoding problem through numerical simulations

    DOI

  • Estimation of the Effects in the Experimental Design Using Fourier Transforms

    Yoshifumi Ukita, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E93A ( 11 ) 2077 - 2082  2010.11  [Refereed]

     View Summary

    We propose that the model in experimental design be ex pressed in terms of an orthonormal system Then we can easily estimate the effects using Fourier transforms We also provide the theorems with respect to the sum of squares needed in analysis of variance Using these theorems It is clear that we can execute the analysis of variance in this model

    DOI

  • A Note on the Branch-and-Cut Approach to Decoding Linear Block Codes

    堀井俊佑, 松嶋敏泰, 平澤茂一

    IEICE Trans. Fundamentals   Vol.E93-A ( No.11 ) 1912 - 1917  2010.11

  • Applying Makov Decision Processes to Selective-Repeat ARQ with Finite Operation Time and Finite Receiver Buffer

    MAEDA Yasunari, AMEMIYA Koji, KOBAYASHI Naoto, YOSHIDA Hideki, SUZUKI Masakiyo, MATSUSHIMA Toshiyasu

    The Transactions of the Institute of Electronics, Information and Communication Engineers. A   93 ( 8 ) 572 - 578  2010.08

    CiNii

  • A Note on a Sampling Theorem for Functions over GF(q)(n) Domain

    Yoshifumi Ukita, Tomohiko Saito, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E93A ( 6 ) 1024 - 1031  2010.06  [Refereed]

     View Summary

    In digital signal processing, the sampling theorem states that any real valued function f can be reconstructed from a sequence of values of f that are discretely sampled with a frequency at least twice as high as the maximum frequency of the spectrum of f. This theorem can also be applied to functions over finite domain. Then, the range of frequencies of f can be expressed in more detail by using a bounded set instead of the maximum frequency. A function whose range of frequencies is confined to a bounded set is referred to as bandlimited function. And a sampling theorem for bandlimited functions over Boolean domain has been obtained. Here, it is important to obtain a sampling theorem for bandlimited functions not only over Boolean domain (GF(2)(n) domain) but also over GF(q)(n) domain, where q is a prime power and GF(q) is Galois field of order q. For example, in experimental designs, although the model can be expressed as a linear combination of the Fourier basis functions and the levels of each factor can be represented by GF(q), the number of levels often take a value greater than two. However, the sampling theorem for bandlimited functions over GF(q)(n) domain has not been obtained. On the other hand, the sampling points are closely related to the codewords of a linear code. However, the relation between the parity check matrix of a linear code and any distinct error vectors has not been obtained, although it is necessary for understanding the meaning of the sampling theorem for bandlimited functions. En this paper, we generalize the sampling theorem for bandlimited functions over Boolean domain to a sampling theorem for bandlimited functions over GF(q)(n) domain. We also present a theorem for the relation between the parity check matrix of a linear code and any distinct error vectors. Lastly, we clarify the relation between the sampling theorem for functions over GF(q)(n) domain and linear codes.

    DOI

  • Optimal Ramp Scheme for Key Predistribution System with Multiple Key Distribution Centers

    YOSHIDA Takahiro, MATSUSHIMA Toshiyasu, IMAI Hideki

    The Transactions of the Institute of Electronics, Information and Communication Engineers. A   93 ( 4 ) 277 - 288  2010.04

    CiNii

  • On the Web Traffic Modeling under the Opening and Closing Services by the Non-Stationary Poisson Process

    Daiki Koizumi, Shunsuke Horii, Toshiyasu Matsushima

    IEICE Technical Report   109 ( 449 ) 343 - 347  2010.03

  • An accurate density evolution analysis for a finite-state Markov channel

    Naoto Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Journal of Discrete Mathematical Sciences and Cryptography   13 ( 1 ) 85 - 97  2010

     View Summary

    In this paper, we consider density evolution analyses of low-density parity-check (LDPC) codes for a finite-state Markov channel (FSMC). Since operations in the subgraph corresponding to the estimation process do not satisfy symmetry conditions, all densities involved in each value of the channel state should be kept track. In [6], to avoid the complexity to compute plural pdfs, only one pdf involved in the channel state is computed exploiting marginalization operations. We suppose that this approach is not accurate enough to track the estimation process of the joint estimation-decoding. In this paper, we present a procedure of the density evolution analysis of LDPC codes for an FSMC of which all densities involved in each value of the channel state are kept track. © 2010 Taylor &amp
    Francis Group, LLC.

    DOI

  • A Linear Programming Bound for Unequal Error Protection Codes

    Tomohiko Saito, Yoshifumi Ukita, Toshiyasu Matsushima, Shigeichi Hirasawa

    2010 AUSTRALIAN COMMUNICATIONS THEORY WORKSHOP     24 - +  2010

     View Summary

    In coding theory, it is important to calculate an upper bound for the size of codes given the length and minimum distance. The Linear Programming (LP) bound is known as a good upper bound for the size of codes. On the other hand, Unequal Error Protection (UEP) codes have been studied in coding theory. In UEP codes, a codeword has special bits which are protected against a greater number of errors than other bits. In this paper, we propose a LP bound for UEP codes. Firstly, we generalize the distance distribution (or weight distribution) of codes. Under the generalization, we lead to the LP bound for UEP codes. And we show a numerical example of the LP bound for UEP codes. Lastly, we compare the proposed bound with a modified Hamming bound.

  • On the Overflow Probability of Fixed-to-Variable Length Codes with Side Information

    Ryo Nomura, Toshiyasu Matsushima

    2010 DATA COMPRESSION CONFERENCE (DCC 2010)     548 - 548  2010

     View Summary

    We consider the source coding problem with side information. Especially, we consider the FV code in the case that the encoder and the decoder can see side information. We obtain the condition that there exists a FV code under the condition that the overflow probability is smaller than or equal to some constant.

    DOI

  • On the Overflow Probability of Lossless Codes with Side Information

    Ryo Nomura, Toshiyasu Matsushima

    2010 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY     131 - 135  2010

     View Summary

    Lossless fixed-to-variable(FV) length codes are considered. The overflow probability is one of criteria that evaluate the performance of FV code. In the single source coding problem, there were many researches on the overflow probability. Recently, the source coding problem for correlated sources, such as Slepian-Wolf coding problem or source coding problem with side information, is one of main topics in information theory. In this paper, we consider the source coding problem with side information. Especially, we consider the FV code in the case that the encoder and the decoder can see side information. In this case, several codes were proposed and their mean code lengths were analyzed. However, there was no research about the overflow probability. We shall show two lemmas about the overflow probability. Then we obtain the condition that there exists a FV code under the condition that the overflow probability is smaller than or equal to some constant.

    DOI

  • On the Appropriate Organizational Size for Each Research Community

    MATSUSHIMA Toshiyasu

    IEICE Fundamentals Review   4 ( 1 ) 2 - 4  2010

    CiNii

  • KL情報量を制約としたResolvability問題における達成可能条件の評価

    野村亮, 吉田隆弘, 松嶋敏泰

    電子情報通信学会論文誌   Vol. J93-A ( No.3 ) 216 - 221  2010

  • 多端子情報理論に基づくセンサネットワークのモデル化と信頼度評価

    野村亮, 松嶋敏泰

    情報処理学会論文誌:数理モデル化と応用   Vol3 ( No.1 ) 13 - 24  2010

  • 複数の鍵配送センターを用いたランプ型鍵事前配布方式

    吉田隆弘, 松嶋敏泰, 今井秀樹

    電子情報通信学会論文誌A   Vol.J93-A ( No.4 ) 277 - 288  2010

  • A note on a sampling theorem for functions over GF(q)n domain

    Yoshifumi Ukita, Tomohiko Saito, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E93-A ( 6 ) 1024 - 1031  2010

     View Summary

    In digital signal processing, the sampling theorem states that any real valued function f can be reconstructed from a sequence of values of f that are discretely sampled with a frequency at least twice as high as the maximum frequency of the spectrum of f . This theorem can also be applied to functions over finite domain. Then, the range of frequencies of f can be expressed in more detail by using a bounded set instead of the maximum frequency. A function whose range of frequencies is confined to a bounded set is referred to as bandlimited function. And a sampling theorem for bandlimited functions over Boolean domain has been obtained. Here, it is important to obtain a sampling theorem for bandlimited functions not only over Boolean domain (GF(2) n domain) but also over GF(q)n domain, where q is a prime power and GF(q) is Galois field of order q. For example, in experimental designs, although the model can be expressed as a linear combination of the Fourier basis functions and the levels of each factor can be represented by GF(q), the number of levels often take a value greater than two. However, the sampling theorem for bandlimited functions over GF(q)n domain has not been obtained. On the other hand, the sampling points are closely related to the codewords of a linear code. However, the relation between the parity check matrix of a linear code and any distinct error vectors has not been obtained, although it is necessary for understanding the meaning of the sampling theorem for bandlimited functions. In this paper, we generalize the sampling theorem for bandlimited functions over Boolean domain to a sampling theorem for bandlimited functions over GF(q)n domain. We also present a theorem for the relation between the parity check matrix of a linear code and any distinct error vectors. Lastly, we clarify the relation between the sampling theorem for functions over GF(q)n domain and linear codes. © 2010 The Institute of Electronics, Information and Communication Engineers.

    DOI

  • マルコフ決定過程の動作時間と受信バッファ容量が有限の選択再送ARQへの適用

    前田康成, 雨宮康二, 小林直人, 吉田秀樹, 鈴木正清, 松嶋敏泰

    電子情報通信学会論文誌A   Vol.J93-A ( No.8 ) 572 - 578  2010

  • A note on automatic construction algorithms for orthogonal designs of experiments using error-correcting codes

    Tomohiko Saito, Toshiyasu Matsushima, Shigeichi Hirasawa

    Journal of Discrete Mathematical Sciences and Cryptography   13 ( 4 ) 369 - 381  2010

     View Summary

    In the field of experimental design, it is important to construct orthogonal designs. In this paper, we propose a new algorithm to construct orthogonal design. This algorithm uses Ukita’s algorithm, which is essentially based on projective geometries, and uses orthogonal designs constructed by error-correcting codes. We show some numerical examples of the proposed algorithm, and show that the proposed algorithm can construct good orthogonal designs with low complexity even if there are high order effects. © 2010 Taylor &amp
    Francis Group, LLC.

    DOI

  • Asymptotic property of universal lossless coding for independent piecewise identically distributed sources

    Tota Suko, Toshiyasu Matsushima, Shigeichi Hirasawa

    Journal of Discrete Mathematical Sciences and Cryptography   13 ( 4 ) 383 - 391  2010

     View Summary

    The universal lossless source coding problem is one of the most important problem in communication systems. The aim of source coding is to compress data to reduce costs in digital communication. Traditional universal source coding schemes are usually designed for stationary sources. Recently, some universal codes for nonstationary sources have been proposed. Independent piecewise identically distributed (i.p.i.d.) sources are simple nonstationary sources that parameter changes discontinuously. In this paper, we assume new i.p.i.d. sources class, and we prove that Bayes codes minimize the mean redundancy when parameter transition pattern is known and parameter is unknown. © 2010 Taylor &amp
    Francis Group, LLC.

    DOI

  • On the Condition of epsilon-Transmissible Joint Source-Channel Coding for General Sources and General Channels

    Ryo Nomura, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E92A ( 11 ) 2936 - 2940  2009.11  [Refereed]

     View Summary

    The joint source-channel coding problem is considered. The joint source-channel coding theorem reveals the existence of a code for the pair of the source and the channel under the condition that the error probability is smaller than or equal to epsilon asymptotically. The separation theorem guarantees that we can achieve the optimal coding performance by using the two-stage coding. In the case that epsilon = 0, Han showed the joint source-channel coding theorem and the separation theorem for general sources and channels. Furthermore the epsilon-coding theorem (0 <= epsilon < 1) in the similar setting was studied. However, the separation theorem was not revealed since it is difficult in general. So we consider the separation theorem in this setting.

    DOI

  • Generalization and Extension of XEX* Mode

    Kazuhiko Minematsu, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E92A ( 2 ) 517 - 524  2009.02  [Refereed]

     View Summary

    This paper describes an extension of XEX* mode, which is a method to convert a block cipher into a tagged tweakable block cipher, a notion introduced by Rogaway in 2004 as an extension of the tweakable block cipher by Liskov et al. Our extension attaches an additional encryption function to the original XEX*, which has some limitation but is slightly faster than the encryption implemented by XEX*. We prove our scheme's security in a general form, where the offset function, a key component of our construction, is not restricted to the one used by XEX*. We also provide some applications of our result, in particular to OCB 2.0, an authenticated encryption based on XEX*.

    DOI

  • Fingerprinting Codes for Multimedia Data against Averaging Attack

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E92A ( 1 ) 207 - 216  2009.01  [Refereed]

     View Summary

    Code construction for digital fingerprinting, which is a copyright protection technique for multimedia, is considered. Digital fingerprinting should deter collusion attacks, where several fingerprinted copies of the same content are mixed to disturb their fingerprints. In this paper, we consider the averaging attack, which is known to be effective for multimedia fingerprinting with the spread spectrum technique. We propose new methods tor constructing fingerprinting codes to increase the coding rate of conventional fingerprinting codes, while they guarantee to identify the same number of colluders. Due to the new fingerprinting codes, the system can deal with a larger number of users to supply digital contents.

    DOI

  • Reducing the Space Complexity of a Bayes Coding Algorithm using an Expanded Context Tree

    Toshiyasu Matsushima, Shigeich Hirasawa

    2009 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, VOLS 1- 4     719 - +  2009

     View Summary

    The context tree models are widely used in a lot of research fields. Patricia[7] like trees are applied to the context trees that are expanded according to the increase of the length of a source sequence in the previous researches of non-predictive source coding and model selection. The space complexity of the Patricia like context trees are O(t) where t is the length of a source sequence. On the other hand, the predictive Bayes source coding algorithm cannot use a Patricia like context tree, because it is difficult to hold and update the posterior probability parameters on a Patricia like tree. So the space complexity of the expanded trees in the predictive Bayes coding algorithm is O(t(2)). In this paper, we propose an efficient predictive Bayes coding algorithm using a new representation of the posterior probability parameters and the compact context tree holding the parameters whose space complexity is O(t).

  • Reducing the Space Complexity of a Bayes Coding Algorithm using an Expanded Context Tree

    Toshiyasu Matsushima, Shigeich Hirasawa

    2009 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, VOLS 1- 4     719 - +  2009

     View Summary

    The context tree models are widely used in a lot of research fields. Patricia[7] like trees are applied to the context trees that are expanded according to the increase of the length of a source sequence in the previous researches of non-predictive source coding and model selection. The space complexity of the Patricia like context trees are O(t) where t is the length of a source sequence. On the other hand, the predictive Bayes source coding algorithm cannot use a Patricia like context tree, because it is difficult to hold and update the posterior probability parameters on a Patricia like tree. So the space complexity of the expanded trees in the predictive Bayes coding algorithm is O(t(2)). In this paper, we propose an efficient predictive Bayes coding algorithm using a new representation of the posterior probability parameters and the compact context tree holding the parameters whose space complexity is O(t).

  • A Note on the Relation between a Sampling Theorem for Functions over a GF(q)(n) Domain and Linear Codes

    Yoshifumi Ukita, Tomohiko Saito, Toshiyasu Matsushima, Shigeichi Hirasawa

    2009 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC 2009), VOLS 1-9     2665 - +  2009

     View Summary

    In this paper, we generalize the sampling theorem for bandlimited functions over the Boolean domain to a sampling theorem for bandlimited functions over a GF(q)(n) domain. We also present a theorem for the relation between the parity check matrix of a linear code and any distinct error vectors. Lastly, we clarify the relation between the sampling theorem for functions over a GF(q)(n) domain and linear codes.

  • On Precise Achievable Conditions in Resolvability Problem Based on the Asymptotic Normality

    野村亮, 松嶋敏泰

    Far East Journal of Electronics and Communications   Vol.3 ( No.3 ) 85 - 99  2009

  • Density Evolution Analysis of Robustness for LDPC Codes over the Gilbert-Elliott Channel

    Manabu Kobayashi, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E91A ( 10 ) 2754 - 2764  2008.10  [Refereed]

     View Summary

    In this paper. we analyze the robustness for low-density parity-check (LDPC) codes over the Gilbert-Elliott (GE) channel. For this purpose we propose a density evolution method for the case where LDPC decoder uses the mismatched parameters for the GE channel. Using this method, we derive the region of tuples of true parameters and mismatched decoding parameters for the GE channel. where the decoding error probability approaches asymptotically to zero.

    DOI

  • A Combined Matrix Ensemble of Low-Density Parity-Check Codes for Correcting a Solid Burst Erasure

    Gou Hosoya, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E91A ( 10 ) 2765 - 2778  2008.10  [Refereed]

     View Summary

    A new ensemble of low-density parity-check (LDPC) codes for correcting a solid burst erasure is proposed. This ensemble is an instance of a combined matrix ensemble obtained by concatenating some LDPC matrices. We derive a new bound on the critical minimum span ratio of stopping sets for the proposed code ensemble by modifying the bound for ordinary code ensemble. By calculating this bound, we show that the critical minimum span ratio of stopping sets for the proposed code ensemble is better than that of the conventional one with keeping the same critical exponent of stopping ratio for both ensemble. Furthermore from experimental results, we show that the average minimum span of stopping sets for a solid burst erasure of the proposed codes is larger than that of the conventional ones.

    DOI

  • Error control codes for parallel channel with correlated errors

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    2008 IEEE INFORMATION THEORY WORKSHOP     421 - +  2008  [Refereed]

     View Summary

    This paper introduces two channel models of correlated parallel channels. Then we analyze structure of error correcting codes over these correlated parallel channels. We derive necessary and sufficient conditions for these codes and some code construction is presented. We also show some upper and lower bounds on the coding rate of the error correcting codes for correlated parallel channels. The introduced channel models are related to burst error channels and the codes analyzed in this paper can be used as asymmetric interleaving codes for burst error channels.

  • An Efficient Design of Irregular LDPC Codes using Beta Approximation for the Gilbert-Elliott Channel

    Manabu Kobayashi, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    2008 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS, VOLS 1-3     1543 - +  2008  [Refereed]

     View Summary

    In this paper, we investigate the design of low-density parity-check (LDPC) codes for the Gilbert-Elliott (GE) channel. Recently, Eckford et al. proposed a design method of irregular LDPC codes using approximate density-evolution (DE) for Markov channels [7]. In the design method proposed by Eckford et al., the probability density function (PDF) of the messages from variable nodes to check nodes is approximated by the Gaussian distribution. In this paper, we first show the method to obtain the accurate PDF of the messages from variable nodes to check nodes by utilizing two DE steps for the Gaussian distribution. We call this method the iterative density approximation (IDA). Using this method, we can design the good LDPC codes. Next, we propose an efficient design method of irregular LDPC codes by using Beta approximation to the PDF of the channel state probability for the GE channel. Consequently, we show that the complexity to calculate PDFs of the channel messages is considerably reduced though the rates of LDPC codes obtained by using the proposed approximation are almost the same as that of the IDA method.

  • A Ramp Scheme for Key Predistribution System against Collusion of Users and Centers

    Takahiro Yoshida, Toshiyasu Matsushima, Hideki Imai

    2008 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS, VOLS 1-3     531 - +  2008

     View Summary

    In this paper, we consider a, ramp scheme for Key Pre-distribution System (KPS). In the ramp scheme, it can be regarded ass one kind of generalization of KPS, and the required resources can be reduced at the cost of a security degradation which depends on the size of users. We define a ramp scheme for KPS, show lower bound on the amount of user's information needed to generate a common key, and design a protocol that realize a ramp scheme for KPS.

  • On the Condition of epsilon-Transmissibility in Joint Source-Channel Coding for Mixed Sources and Mixed Channels

    Ryo Nomura, Toshiyasu Matsushima, Shigeichi Hirasawa

    2008 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS, VOLS 1-3     1299 - +  2008

     View Summary

    In this study, we consider a joint source-channel coding problem. The joint source-channel coding theorem reveals the existence of a code for the pair of the source V and the channel W under the condition that the probability of error is smaller than or equal to E asymptotically. We show the separation theorem for general sources and general channels.

  • 拡張された有本-Blahutアルゴリズムの大域的収束性について

    安井謙介, 須子統太, 松嶋敏泰

    電子情報通信学会論文誌   Vol.91-A ( No.9 ) 846 - 860  2008

  • 外れ値データの発生を含む回帰モデルに対するベイズ予測アルゴリズム

    須子統太, 松嶋敏泰, 平澤茂一

    情報処理学会論文誌数理モデル化と応用   Vol.1 ( No.1 ) 17 - 26  2008

  • Improved MACs from differentially-uniform permutations

    Kazuhiko Minematsu, Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E90A ( 12 ) 2908 - 2915  2007.12  [Refereed]

     View Summary

    This paper presents MACs that combine a block cipher and its component such as a reduced-round version. Our MACs are faster than the standard MAC modes such as CBC-MAC, and provably secure if the block cipher is pseudorandom and its component is a permutation with a small differential probability. Such a MAC scheme was recently proposed by one of authors, and we provide improvements about security and treading-off between speed and amount of preprocessing.

    DOI

  • Reliability-based hybrid ARQ scheme with encoded parity bit retransmissions and message passing decoding

    Daiki Koizumi, Naoto Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E90A ( 9 ) 1736 - 1744  2007.09  [Refereed]

     View Summary

    Reliability-based hybrid ARQ (RB-HARQ) is a kind of incremental-redundancy ARQ recently introduced. In the RB-HARQ, the receiver returns both NAK signal and set of unreliable bit indices if the received sequence is not accepted. Since each unreliable bit index is determined by the bitwise posterior probability, better approximation of that probability becomes crucial as the number of retransmissions increases. Assuming the systematic code for the initial transmission, the proposed RB-HARQ scheme has the following two features: (a) the sender retransmits newly encoded and interleaved parity bits corresponding to the unreliable information bits; (b) the retransmitted parity bits as well as the initial received sequence are put all together to perform the message passing decoding i.e. the suboptimal MAP decoding. Finally, simulation results are shown to evaluate the above two features.

    DOI

  • Improved collusion-secure codes for digital fingerprinting based on finite geometries

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    2007 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS, VOLS 1-8     522 - +  2007  [Refereed]

     View Summary

    Digital fingerprinting, a copyright protection technique for digital contents, is considered. Digital fingerprinting should deter collusion attacks, where several fingerprinted copies of the same content are mixed to disturb their fingerprints. In this paper, we consider the averaging attack, which has effect for multimedia fingerprinting. We propose new collusion-secure fingerprinting codes based on finite geometries (FGs) which increase the rate of conventional collusion-secure codes, while they guarantee to identify the same number of colluders. Due to the new FG-based fingerprinting codes, the system can deal with a larger number of users to distribute a digital content.

  • Word segmentation for the sequences emitted from a word-valued source

    Takashi Ishida, Toshiyasu Matsushima, Shigeichi Hirasawa

    2007 CIT: 7TH IEEE INTERNATIONAL CONFERENCE ON COMPUTER AND INFORMATION TECHNOLOGY, PROCEEDINGS     662 - 667  2007  [Refereed]

     View Summary

    Word segmentation is the most fundamental and important process for Japanese or Chinese language processing. Because there is no separation between words in these languages, we firstly have to separate the sequence into words. On this problem, it is known that the approach by probabilistic language model is highly efficient, and this is shown practically. On the other hand, recently, a word-valued source has been proposed as a new class of source model for the source coding problem. This model can be supposed to reflect more of the probability structure of natural languages. We may regard Japanese sentence or Chinese sentence as the sequence emitting from a non-prefix-free WVS. In this paper as the first phase of applying WVS to natural language processing, we formulate a word segmentation problem for the sequence from non-prefix-free WVS. Then, we examine the performance of word segmentation for the models by numerical computations.

  • A note on morphological analysis methods based on statistical decision theory

    Yasunari Maeda, Naoya Ikeda, Hideki Yoshida, Yoshitaka Fujiwara, Toshiyasu Matsushima

    PROCEEDINGS OF SICE ANNUAL CONFERENCE, VOLS 1-8     1558 - +  2007  [Refereed]

     View Summary

    Morphological analysis is one of important topics in the field of NLP(Natural Language Processing). In many previous research a HMM(Hidden Markov Model) with unknown parameters has been used as a language model. In this research we also use the HMM as the language model. And we assume that sate transitions in the HMM are dominated by a second order Markov chain. At first we propose two types of morphological analysis methods which minimize the error rate with reference to a Bayes criterion. But the computational complexity of the proposed Bayes optimal morphological analysis methods are exponential order. So we also propose approximate methods.

  • Text Data Compression by Bayes Coding Algorithm

    Akira Nakano, Daiki Koizumi, Toshiyasu Matsushima

    IPSJ SIG Notes   2007-AL-110 ( 5 ) 15 - 20  2007.01

  • Tweakable enciphering schemes from hash-sum-expansion

    Kazuhiko Minematsu, Toshiyasu Matsushima

    PROGRESS IN CRYPTOLOGY - INDOCRYPT 2007   4859   252 - 267  2007

     View Summary

    We study a tweakable blockcipher for arbitrarily long message (also called a tweakable enciphering scheme) that consists of a universal hash function and an expansion, a keyed function with short input and long output. Such schemes, called HCTR and HCH, have been recently proposed. They used (a variant of) the counter mode of a blockcipher for the expansion. We provide a security proof of a structure that underlies HCTR and HCH. We prove that the expansion can be instantiated with any function secure against Known-plaintext attacks (KPAs), which is called a weak pseudorandom function (WPRF). As an application of our proof, we provide efficient blockcipher-based schemes comparable to HCH and HCTR. For the double-block-length case, our result is an interesting extension of previous attempts to build a double-block-length cryptographic permutation using WPRF.

  • A class of prior distributions on context tree models and an efficient algorithm of the Bayes codes assuming it

    Toshiyasu Matsushima, Shigeicb Hirasawa

    2007 IEEE INTERNATIONAL SYMPOSIUM ON SIGNAL PROCESSING AND INFORMATION TECHNOLOGY, VOLS 1-3     305 - +  2007

     View Summary

    The CTW(Context Tree Weighting) algorithm is an efficient universal coding algorithm on context tree models. The CTW algorithm has been interpreted as the non-predictive Bayes coding algorithm assuming a special prior distribution over context tree models. An efficient recursive calculation method using a gathering context tree in the CTW algorithm is well known. Although there exist efficient recursive algorithms for the Bayes codes assuming a special class of prior distributions, the basic property of the prior distribution class has been scarcely investigated. In this paper we show the exact definition of a prior distribution class on context tree models that has the similar property to the class of conjugate priors. We show the posterior distribution is also included in the same distribution class as the prior distribution class. So we can also construct an efficient algorithm of predictive Bayes codes on context tree models by using the prior distribution class. Lastly the asymptotic mean code length of the codes is investigated.

  • A note on error correction schemes using LDPC codes with a high-capacity feedback channel

    Naoto Kobayashi, Toshiyasu Matsushima, Shigeich Hirasawa

    2007 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS, VOLS 1-7     2381 - 2385  2007

     View Summary

    In this paper, transmission schemes with noiseless and high capacity feedback channel is considered. We propose two types of transmission schemes using LDPC codes and clarify the density evolution analysis method for these proposed schemes. We investigate the performance of the proposed schemes by the density evolution analysis and computer simulations. The result shows some interesting characteristics for schemes with high capacity feedback channel.

  • A note on error correction schemes using LDPC codes with a high-capacity feedback channel

    Naoto Kobayashi, Toshiyasu Matsushima, Shigeich Hirasawa

    2007 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS, VOLS 1-7     2381 - 2385  2007

     View Summary

    In this paper, transmission schemes with noiseless and high capacity feedback channel is considered. We propose two types of transmission schemes using LDPC codes and clarify the density evolution analysis method for these proposed schemes. We investigate the performance of the proposed schemes by the density evolution analysis and computer simulations. The result shows some interesting characteristics for schemes with high capacity feedback channel.

  • An algorithm for computing the secrecy capacity of broadcast channels with confidential messages

    Kensuke Yasui, Tota Suko, Toshiyasu Matsushima

    2007 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS, VOLS 1-7     936 - 940  2007

     View Summary

    In this paper, we present an iterative algorithm for computing the secrecy capacity of broadcast channel with confidential message (BCC) in the situation that the main channel is less noisy than the eavesdropper's channel. The global convergence of the algorithm is proved, and an expression for its convergence rate is derived.

  • On the epsilon-overflow probability of lossless codes

    Ryo Nomura, Toshiyasu Matsushima, Shigeichi Hirasawa

    2007 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS, VOLS 1-7     441 - +  2007

     View Summary

    In this paper, we generalize the achievability of variable-length coding from two viewpoints. One is the definition of an overflow probability, and the other is the definition of an achievability. We define the overflow probability as the probability of codeword length, not per symbol, is larger than eta(n) and we introduce the epsilon-achievability of variable-length codes that implies an existence of a code for the source under the condition that the overflow probability is smaller than or equal to epsilon.
    Then we show that the epsilon-achievabitity of variable-length codes is essentially equivalent to the epsilon-achievability of fixed-length codes for general sources. Moreover we show the condition of epsilon-achievability for some restricted sources given epsilon.

    DOI

  • ストリーム暗号における擬似乱数生成器の構成に関する一考察

    三上暢仁, 齋藤友彦, 松嶋敏泰

    電子情報通信学会論文誌A   Vol.J90-A ( No.5 ) 470 - 476  2007

  • 無ひずみ情報源符号化におけるオーバフロー確率について

    野村亮, 松嶋敏泰, 平澤茂一

    電子情報通信学会論文誌   Vol.J90-A ( No.4 ) 292 - 304  2007

  • A note on the ∈-overflow probability of lossless codes

    Ryo Nomura, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E90-A ( 12 ) 2965 - 2970  2007

     View Summary

    In this letter, we generalize the achievability of variable-length coding from two viewpoints. One is the definition of an overflow probability, and the other is the definition of an achievability. We define the overflow probability as the probability of codeword length, not per symbol, is larger than ηn and we introduce the e-achievability of variable-length codes that implies an existence of a code for the source under the condition that the overflow probability is smaller than or equal to ∈. Then we show that the e-achievability of variable-length codes is essentially equivalent to the e-achievability of fixed-length codes for general sources. Moreover by using above results, we show the condition of e-achievability for some restricted sources given ∈. Copyright © 2007 The Institute of Electronics, Information and Communication Engineers.

    DOI

  • Properties of a word-valued source with a non-prefix-free word set

    Takashi Ishida, Masayuki Goto, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E89A ( 12 ) 3710 - 3723  2006.12  [Refereed]

     View Summary

    Recently, a word-valued source has been proposed as a new class of information source models. A word-valued source is regarded as a source with a probability distribution over a word set. Although a word-valued source is a nonstationary source in general, it has been proved that an entropy rate of the source exists and the Asymptotic Equipartition Property (AEP) holds when the word set of the source is prefix-free. However, when the word set is not prefix-free (non-prefix-free), only an upper bound on the entropy density rate for an i.i.d. word-valued source has been derived so far. In this paper, we newly derive a lower bound on the entropy density rate for an i.i.d. word-valued source with a finite non-prefix-free word set. Then some numerical examples are given in order to investigate the behavior of the bounds.

    DOI

  • A note on error correction schemes with a feedback channel

    Naoto Kobayashi, Daiki Koizumi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E89A ( 10 ) 2475 - 2480  2006.10  [Refereed]

     View Summary

    We propose a new fixed-rate error correction system with a feedback channel. In our system, the receiver transmits a list of positions of unreliable information bits based on the log a-posteriori probability ratios by outputs of a soft-output decoder to the transmitter. This method is just like that of the reliability-based hybrid ARQ scheme. To dynamically select an appropriate interleaving function with feedback information is a key feature of our system. By computer simulations, we show that the performance of a system with a feedback channel is improved by dynamically selecting an appropriate interleaving function.

    DOI

  • A Study of Bayes Coding for i.i.d. Sources with Consideration of the Generating Patterns of the Symbols in the Source Alphabet

    Ryunosuke Nanmo, Daiki Koizumi, Toshiyasu Matsushima

    IEICE Technical Report   106 ( 184 ) 25 - 30  2006.07

  • The Reliability based Hybrid ARQ Scheme with both the Encoded Parity Bit Retransmissions and Message Passing Decoding

    Daiki Koizumi, Naoto Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Technical Report of IEICE (IT2006-17)   106 ( 60 ) 23 - 28  2006.05

  • A note on construction of orthogonal arrays with unequal strength from error-correcting codes

    Tomohiko Saito, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E89A ( 5 ) 1307 - 1315  2006.05  [Refereed]

     View Summary

    Orthogonal Arrays (OAs) have been playing important roles in the field of experimental design. It has been known that OAs are closely related to error-correcting codes. Therefore, many OAs can be constructed from error-correcting codes. But these OAs are suitable for only cases that equal interaction effects can be assumed, for example, all two-factor interaction effects. Since these cases are rare in experimental design, we cannot say that OAs from error-correcting codes are practical. In this paper. we define OAs with unequal strength. In terms of our terminology. OAs from error-correcting codes are OAs with equal strength. We show that OAs with unequal strength are closer to practical OAs than OAs with equal strength. And we clarify the relation between OAs with unequal strength and unequal error-correcting codes. Finally, we propose some construction methods of OAs with unequal strength from unequal error-correcting codes.

    DOI

  • Transformation of a parity-check matrix for a message-passing algorithm over the BEC

    Naoto Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E89A ( 5 ) 1299 - 1306  2006.05  [Refereed]

     View Summary

    We propose transformation of a parity-check matrix of any low-density parity-check code. A code with transformed parity-check matrix is an equivalent of a code with the original parity-check matrix. For the binary erasure channel, performance of a message-passing algorithm with a transformed parity-check matrix is better than that with the original matrix.

    DOI

  • Adaptive incremental redundancy coding and decoding schemes using feedback information

    Toshiyasu Matsushima

    Proceedings of 2006 IEEE Information Theory Workshop     189 - 193  2006

     View Summary

    If we can use a high capacity feedback channel, the receiver may transmit the indices of unreliable bits in a received sequence instead of ACK/NAK signals. We consider the error correction schemes using the indices of unreliable bits. The suitable parity symbols are incrementally transmitted using the feedback information. The bitwise reliability can be calculated by message passing algorithms on the variable graph constructed by adding the new nodes and arcs corresponding to the received parity bits to a former graph at each stage.

  • On the Interleaver Design Method for Block Turbo Codes and Its Minimum Distance

    Manabu Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

      vol.J89-A ( no.2 ) 129 - 143  2006  [Refereed]

  • On Improvement of Error Exponents for Decision Feedback Scheme with LR+Th

    Toshihiro Niinomi, Toshiyasu Matsushima, Shigeichi Hirasawa

      vol.J89-A ( no.12 ) 1168 - 1174  2006  [Refereed]

  • ストリーム暗号への攻撃法の改良に関する一考察-多次元の相関を利用した攻撃-

    細渕智史, 斉藤友彦, 松嶋敏泰

    電子情報通信学会論文誌   Vol.J89-A ( No.2 ) 121 - 128  2006

  • A Note on HTTP Traffic Analysis of the Time Series Model with a Time Varying Density Parameter

    Daiki Koizumi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. of the 28th Symposium on Information Theory and Its Applications (SITA2005)   2   729 - 732  2005.12

  • 単一ループをもつグラフィカルモデルにおける確率伝播型アルゴリズムに関する一考察

    岡野 洋平, 小泉 大城, 松嶋 敏泰

    第28回情報理論とその応用学会(SITA2005)予稿集   1   1 - 4  2005.12

  • A heuristic search method with the reduced list of test error patterns for maximum likelihood decoding

    H Yagi, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E88A ( 10 ) 2721 - 2733  2005.10  [Refereed]

     View Summary

    The reliability-based heuristic search methods for maximum likelihood decoding (MLD) generate test error patterns (or, equivalently, candidate codewords) according to their heuristic values. Test error patterns are stored in lists and its space complexity is crucially large for MLD of long block codes. Based on the decoding algorithms both of Battail and Fang and of its generalized version suggested by Valembois and Fossorier, we propose a new method for reducing the space complexity of the heuristic search methods for MLD including the well-known decoding algorithm of Han et al. If the heuristic function satisfies a certain condition, the proposed method guarantees to reduce the space complexity of both the Battail-Fang and Han et al. decoding algorithms. Simulation results show the high efficiency of the proposed method.

    DOI

  • A Study of Reliability Based Hybrid ARQ Schemes Using a Recursive Systematic Convolutional Code

    Naoto Kobayashi, Daiki Koizumi, Toshiyasu Matsushima, Shigeichi Hirasawa

      105 ( 311 ) 7 - 11  2005.09

  • インターネットトラヒックのポアソン分布に従う非定常な時系列モデルを用いた解析に関する一考察

    小泉 大城, 松嶋 敏泰, 平澤 茂一

    第4回情報科学技術フォーラム(FIT2005)講演論文集   4   35 - 38  2005.08

  • A Study of Reliability Based Hybrid ARQ Scheme with Bitwise Posterior Probability Evaluation from Message Passing Algorithm

    Daiki Koizumi, Naoto Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Technical Report of IEICE (IT2005-24)   105 ( 85 ) 11 - 16  2005.05

  • Bayes universal coding algorithm for side information context tree models

    T Matsushima, S Hirasawa

    2005 IEEE International Symposium on Information Theory (ISIT), Vols 1 and 2     2345 - 2348  2005

     View Summary

    The problem of universal codes with side information is investigated from Bayes criterion. We propose side information context tree models which are an extension of context tree models to sources with side information. Assuming a special class of the prior distributions for side information context tree models, we propose an efficient algorithm of Bayes code for the models. The asymptotic code length of the Bayes codes with side information is also investigated.

  • A note on a decoding algorithm of codes on graphs with small loops

    N Kobayashi, T Matsushima, S Hirasawa

    Proceedings of the IEEE ITSOC Information Theory Workshop 2005 on Coding and Complexity     109 - 113  2005

     View Summary

    The best-known algorithm for the decoding of low-density parity-check (LDPC) codes is the sum-product algorithm (SPA). The SPA is a message-passing algorithm on a graphical model called a factor graph (FG). The performance of the SPA depends on a structure of loops in a FG. Pearl showed that loops in a graphical model could be erased by the clustering method. This method clusters plural nodes into a single node. In this paper, we show several examples about a decoding on a FG to which the clustering method is applied. And we propose an efficient decoding algorithm for it.

  • On Error Exponents for Variable Size List Decoder Using the Viterbi Algorithm with Likelihood Ratio Testing

    Toshihiro Niinomi, Toshiyasu Matsushima, Shigeichi Hirasawa

      vol.J88-A ( no.11 ) 1343 - 1351  2005  [Refereed]

  • A Classification of the Probabilistic Reasoning given Distribution Evidence and Kullback-Leibler Information

    Toshiyasu Matsushima

        163 - 173  2005

     View Summary

    The probabilistic reasoning given distribution evidence, virtual evidence, indirect evidence or likelihood have been investigated in previous research.In this paper, we classify the reasoning into two types and define each type of the reasoning by mathematical formulas. We clarify the difference between two types by the definition. From the definition, we show that the first type of reasoning is solved by the minimization of Kullback-Leibler(K-L) information under marginal constraints and the second type is calculated by the ordinary probabilistic reasoning methods such as Belief Propagation(BP). We also show that the Iterative Scaling Procedure(ISP) is applied to the first type reasoning. Moreover, we propose an efficient propagation algorithm, which are based on ISP, for the reasoning on Junction trees. Both the space and the time complexities of the proposed algorithm are lower than that of the previous research

  • ターボ符号・LDPC符号とその復号法の概要

    松嶋敏泰

    電子情報通信学会誌   Vol.88 ( No.4 ) 244 - 248  2005

  • 質問学習に直交計画を用いた場合の予測アルゴリズムに関する一考察

    浮田 善文, 小泉 大城, 松嶋 敏泰

    第57回人工知能基本問題研究会(SIG-FPAI)     25 - 32  2004.11

  • Efficient Reliability-based Soft Decision Decoding Algorithm over Markov Modulated Channel

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Trans. Fundamentals     823 - 828  2004.08

  • An Improvement Method of Reliability-Based Maximum Likelihood Decoding Algorithms Using an Order Relation among Binary Vectors

    Hideki Yagi, Nanabu Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Trans. Fundamentals     2493 - 2502  2004.08

  • Tailbiting 畳込み符号の復号アルゴリズムに関する一考察

    岡野 洋平, 小泉 大城, 松嶋 敏泰

    電子情報通信学会 技術研究報告(IT2004-26)   104 ( 229 ) 47 - 52  2004.07

  • A New Decoding Algorithm Using Likelihood Ratio Testing for Tree Codes

    Toshihiro Niinomi, Toshiyasu Matsushima, Shigeichi Hirasawa

      vol.J87-A ( no.2 ) 224 - 233  2004  [Refereed]

  • A source model with probability distribution over word set and recurrence time theorem

    M Goto, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E86A ( 10 ) 2517 - 2525  2003.10  [Refereed]

     View Summary

    Nishiara and Morita defined an i.i.d. word-valued source which is defined as a pair of an i.i.d. source with a countable alphabet and a function which transforms each symbol into a word over finite alphabet. They showed the asymptotic equipartition property (AEP) of the i.i.d. word-valued source and discussed the relation with source coding algorithm based on a string parsing approach. However, their model is restricted in the i.i.d. case and any universal code for a class of word-valued sources isn't discussed. In this paper, we generalize the i.i.d. word-valued source to the ergodic word-valued source which is defined by an ergodic source with a countable alphabet and a function from each symbol to a word. We show existence of entropy rate of the ergodic word-valued source and its formula. Moreover, we show the recurrence time theorem for the ergodic word-valued source with a finite alphabet. This result clarifies that Ziv-Lempel code (ZL77 code) is universal for the ergodic word-valued source.

  • 直交計画を用いたブール関数の学習に関する一考察

    浮田善文, 松嶋敏泰, 平澤茂一

    電子情報通信学会論文誌   J86-A ( 4 ) 482 - 490  2003.04

  • 情報理論からの学習,帰納推論,予測 (特集 情報論的学習理論--機械学習のさまざまな形)

    松嶋敏泰

    Computer today   20 ( 2 ) 5 - 10  2003.03

  • A parallel iterative decoding algorithm for zero-tail and tail-biting convolutional codes

    T Matsushima, TK Matsushima, S Hirasawa

    2003 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY - PROCEEDINGS     175 - 175  2003

  • A parallel iterative decoding algorithm for zero-tail and tail-biting convolutional codes

    T Matsushima, TK Matsushima, S Hirasawa

    2003 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY - PROCEEDINGS     175 - 175  2003

  • メモリ量を低減した近似ベイズ符号化アルゴリズム

    野村亮, 松嶋敏泰, 平澤茂一

    電子情報通信学会論文誌   J86-A ( 1 ) 46 - 59  2003.01

  • ウェーブレットパケット基底を用いた信号推定におけるベイズ決定理論の適用に関する一考察

    北原正樹, 野村亮, 松嶋敏泰

    電子情報通信学会論文誌(A) Vol.J85   J85-A ( 5 ) 584 - 596  2002.05

  • An alternate algorithm for calculating generalized posterior probability and decoding

    T Matsushima, TK Matsushima, S Hirasawa

    ISIT: 2002 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS     338 - 338  2002

  • On the generalized viterbi algorithm using likelihood ratio testing

    T Niinomi, T Matsushima, S Hirasawa

    ISIT: 2002 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS     366 - 366  2002

  • An analysis of the difference of code lengths between two-step codes based on MDL principle and bayes codes

    M Goto, T Matsushima, S Hirasawa

    IEEE TRANSACTIONS ON INFORMATION THEORY   47 ( 3 ) 927 - 944  2001.03  [Refereed]

     View Summary

    In this paper, we discuss the difference in code lengths between the code based on the minimum description length (MDL) principle (the MDL code) and the Bayes code under the condition that the same prior distribution is assumed for both codes. It is proved that the code length of the Bayes code is smaller than that of the MDL code by o(1) or O(1) for the discrete model class and by O(1) for the parametric model class. Because we can assume the same prior for the Bayes code as for the code based on the MDL principle, it is possible to construct the Bayes code with equal or smaller code length than the code based on the MDL principle. From the viewpoint of mean code length per symbol unit (compression rate), the Bayes code is asymptotically indistinguishable from the MDL two-stage codes.

  • 情報論的学習理論における数理モデル

    松嶋敏泰

    人工知能学会誌   16   252 - 255  2001.03

  • A study of the decision of control parameters for adaptive automatic-repeat request strategy

    K Tsukamoto, T Matsushima, S Hirasawa

    ELECTRONICS AND COMMUNICATIONS IN JAPAN PART I-COMMUNICATIONS   84 ( 11 ) 61 - 70  2001  [Refereed]

     View Summary

    Ill many practical transmission systems, transmission error rate of the channel is time-varying. In previous works for controlling the rime-varying channel, the state of the channel is estimated and is uniquely determined. Then, the optimum value is chosen as a control parameter for the determined state. However, there is a discrepancy between the actual channel state and estimated state. The optimum control parameter determined for the estimated state does not necessarily provide the best performance presented by some object function in error control such as throughput efficiency. In th is study, the expected value of throughput is calculated by using the posterior probability for each state and an algorithm in which the control parameter is selected to maximize the expected performance is proposed. From the viewpoint of the statistical decision theory, it is shown that this selection is the optimum in maximizing the throughput under the Bayes criterion. Using simulation, the proposed method is evaluated and its effectiveness is shown. (C) 2001 Scripta Technica

  • Mathematical Models in Information-Based Induction Science

    Toshiyasu Matsushima

    Journal of the Japanese Society for Artificial Intelligence   vol.16 ( no.2 ) 252 - 255  2001  [Refereed]

  • 損失関数を考慮した拡張事後密度の漸近正規性

    後藤正幸, 松嶋敏泰, 平澤茂一

    電子情報通信学会論文誌   J83-A ( 6 ) 639 - 650  2000.06

  • ベイズ決定理論に基づく統計的モデル選択の選択誤り確率に関する解析

    日本経営工学会   50 ( No.6 ) 475 - 485  2000.02

  • On Analysis of noiseless decision feedback scheme using fixed size list decoder for tree codes

    T Niinomi, T Matsushima, S Hirasawa

    2000 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS     231 - 231  2000

     View Summary

    In this paper, the generalized proof are shown for the coding theorem of [1]. Consequently, the further discussion is obtained.

  • Achievable rates of random number generators for an arbitrary prescribed distribution from an arbitrary given distribution

    T Yoshida, T Matsushima, S Hirasawa

    2000 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS     155 - 155  2000

     View Summary

    In this paper, we show maximal rates in the case that random number generators generate a random sequence with an arbitrary prescribed distribution from a random sequence with an arbitrary given distribution.

  • On the variance and the probability of length overflow of lossless codes

    R Nomura, T Matsushima, S Hirasawa

    2000 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS     44 - 44  2000

     View Summary

    In this paper, we show the probability of length overflow of several codes by using the variance and the asymptotic normality of the codelength.

  • 木符号におけるリスト復号法を用いた判定帰還方式について

    電子情報通信学会論文誌/電子情報通信学会   J83-A ( No.1 ) 67 - 82  2000.01

  • 不確実な知識を用いた推論のモデル化と推論法について

    情報処理学会論文誌/情報処理学会   41 ( No.1 ) 1 - 11  2000.01

  • Almost sure and mean convergence of extended stochastic complexity

    M Gotoh, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E82A ( 10 ) 2129 - 2137  1999.10  [Refereed]

     View Summary

    We analyse the extended stochastic complexity (ESC) which has been proposed by K. Yamanishi. The ESC can be applied to learning algorithms for on-line prediction and batch-learning settings. Yamanishi derived the upper bound of ESC satisfying uniformly for all data sequences and that of the asymptotic expectation of ESC. However, Yamanishi concentrates mainly on the worst case performance and the lower bound has not been derived. In this paper, we show some interesting properties of ESC which are similar to Bayesian statistics: the Bayes rule and the asymptotic normality. We then derive the asymptotic formula of ESC in the meaning of almost sure and mean convergence within an error of o(1) using these properties.

  • On decoding methods beyond the BCH bound and their applications to soft-decision decoding

    Manabu Kobayashi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Electronics and Communications in Japan, Part III: Fundamental Electronic Science (English translation of Denshi Tsushin Gakkai Ronbunshi)   82 ( 9 ) 39 - 51  1999.09  [Refereed]

     View Summary

    For the two-dimensional BCH code, several decoding methods exceeding the BCH bound and correcting the errors that cannot be corrected by the conventional limited distance decoding method have been proposed. This article proposes an algorithm that allows reduction of the computational volume in a decoding method exceeding the BCH bound by solving the equation for unknown variables beforehand and limiting the range of the error location. Further, this decoding method exceeding the BCH bound is applied to soft-decision decoding methods that use limited-distance decoding multiple times, and especially to Chase decoding, the decoding of Tanaka et al., and that of Kaneko et al. It is shown that the amount of computation and the decoding error rate are improved.

    DOI

  • 電子情報通信用語辞典

    コロナ社    1999.07

  • The Generalization of Bayesian Network's Deductive Method

    Daiki Koizumi, Yoshifumi Ukita, Toshiyasu Matsushima

    The 1999 Sigma Xi, The Scientific Research Society, Florida Tech Chapter    1999.04

  • 事前分布が異なる場合のMDL原理に基づく符号とベイズ符号の符号長に関する解析

    電子情報通信学会論文誌   J82-A ( 5 ) 698 - 708  1999

  • マルコフ情報源に対し誤り伝播を抑えるVF符号に関する一考察

    電子情報通信学会論文誌   J82-A ( 5 ) 736 - 741  1999

  • BCH限界を超える復号アルゴリズムを用いた2元BCH符号の軟判定復号法

    電子情報通信学会論文誌   J82-A ( 4 ) 539 - 549  1999

  • A generalization of B.S. Clarke and A.R. Barron's asymptotics of Bayes codes for FSMX sources

    M Gotoh, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E81A ( 10 ) 2123 - 2132  1998.10  [Refereed]

     View Summary

    We shall generalize B.S. Clarke and A.R. Barron's analysis of the Bayes method for the FSMX sources. The FSMX source considered here is specified by the set of all states and its parameter value. At first, we show the asymptotic codelengths of individual sequences of the Bayes codes for the FSMX sources. Secondly, we show the asymptotic expected codelengths. The Bayesian posterior density and the maximum likelihood estimator satisfy asymptotic normality for the finite ergodic Markov source, and this is the key of our analysis.

  • A generalization of B.S. Clarke and A.R. Barron's asymptotics of Bayes codes for FSMX sources

    M Gotoh, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E81A ( 10 ) 2123 - 2132  1998.10  [Refereed]

     View Summary

    We shall generalize B.S. Clarke and A.R. Barron's analysis of the Bayes method for the FSMX sources. The FSMX source considered here is specified by the set of all states and its parameter value. At first, we show the asymptotic codelengths of individual sequences of the Bayes codes for the FSMX sources. Secondly, we show the asymptotic expected codelengths. The Bayesian posterior density and the maximum likelihood estimator satisfy asymptotic normality for the finite ergodic Markov source, and this is the key of our analysis.

  • Parallel architecture for generalized LFSR in LSI built-in self testing

    TK Matsushima, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E81A ( 6 ) 1252 - 1261  1998.06  [Refereed]

     View Summary

    This paper presents a new architecture for multiple-input signature analyzers. The proposed signature analyzer with H delta inputs is designed by parallelizing a GLFSR(delta, m), where delta is the number of input signals and m is the number of stages in the feedback shift register. The GLFSR, developed by Pradhan and Gupta, is a general framework for representing LFSR-based signature analyzers. The parallelization technique described in this paper can be applied to any kind of GLFSR signature analyzer, e.g., SISRs, MISRs, multiple MISRs and MLFSRs. It is shown that a proposed signature analyzer with H delta inputs requires less complex hardware than either single GLFSR(H delta, m)s or a parallel construction of the H original GLFSR(delta, m)s. It is also shown that the proposed signature analyzer, while requiring simpler hardware, has comparable aliasing probability with analyzers using conventional GLFSRs for some CUT error models of the same test response length and test lime. The proposed technique would be practical for testing CUTs with a large number of output sequences, since the test circuit occupies a smaller area on the LSI chip than the conventional multiple-input signature analyzers of comparable aliasing probability.

  • Parallel architecture for generalized LFSR in LSI built-in self testing

    TK Matsushima, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E81A ( 6 ) 1252 - 1261  1998.06  [Refereed]

     View Summary

    This paper presents a new architecture for multiple-input signature analyzers. The proposed signature analyzer with H delta inputs is designed by parallelizing a GLFSR(delta, m), where delta is the number of input signals and m is the number of stages in the feedback shift register. The GLFSR, developed by Pradhan and Gupta, is a general framework for representing LFSR-based signature analyzers. The parallelization technique described in this paper can be applied to any kind of GLFSR signature analyzer, e.g., SISRs, MISRs, multiple MISRs and MLFSRs. It is shown that a proposed signature analyzer with H delta inputs requires less complex hardware than either single GLFSR(H delta, m)s or a parallel construction of the H original GLFSR(delta, m)s. It is also shown that the proposed signature analyzer, while requiring simpler hardware, has comparable aliasing probability with analyzers using conventional GLFSRs for some CUT error models of the same test response length and test lime. The proposed technique would be practical for testing CUTs with a large number of output sequences, since the test circuit occupies a smaller area on the LSI chip than the conventional multiple-input signature analyzers of comparable aliasing probability.

  • 矛盾を含む知識の取り扱いについての一考察

    人工知能学会誌   13 ( 2 ) 252 - 262  1998.03

  • Type for hierarchical probability model class and universal coding

    T Matsushima, S Hirasawa

    1998 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY - PROCEEDINGS     319 - 319  1998

  • On error rates of statistical model selection based on information criteria

    H Gotoh, T Matsushima, S Hirasawa

    1998 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY - PROCEEDINGS     417 - 417  1998

     View Summary

    In this paper, we shall derive the upper bounds on error rates of the statistical model selection using the information criteria, P(e)*. The similar bounds were derived by J.Suzuki [2]. We shall generalize the results for the general model class.

  • Type for hierarchical probability model class and universal coding

    T Matsushima, S Hirasawa

    1998 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY - PROCEEDINGS     319 - 319  1998

  • On error rates of statistical model selection based on information criteria

    H Gotoh, T Matsushima, S Hirasawa

    1998 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY - PROCEEDINGS     417 - 417  1998

     View Summary

    In this paper, we shall derive the upper bounds on error rates of the statistical model selection using the information criteria, P(e)*. The similar bounds were derived by J.Suzuki [2]. We shall generalize the results for the general model class.

  • The Optimal Algorithms for the Reinforcement Learning Problem Separated into a Learning Period and a Contorol Period

      39 ( 4 ) 1116 - 1126  1998

  • A Note on Decision Theoretic Formulation for Learning from Queries

      39 ( 11 ) 2937 - 2948  1998

  • 適応型ARQにおける制御パラメータ決定方式について

    電子情報通信学会論文誌   J81-B-1 ( 6 ) 391 - 400  1998

  • 質問からの学習問題の決定理論による定式化に関する一考察

    情報処理学会論文誌   39 ( 11 ) 2937 - 2948  1998

  • 学習期間と制御期間に分割された強化学習問題における最適アルゴリズムの提案(共著)

    情報処理学会論文誌   39 ( 4 ) 1116 - 1126  1998

  • BCH限界を超える復号法とその軟判定復号法への応用(共著)

    電子情報通信学会論文誌   J81-A ( 4 ) 751 - 762  1998

  • Berlekamp-Masseyアルゴリズムを用いたBCH限界を超える復号法の計算量について

    電子情報通信学会論文誌A   vol.J80 ( 9 ) 1554 - 1558  1997.09

  • Machine learning by a subset of hypotheses

    T Mukouchi, T Matsushima, S Hirasawa

    SMC '97 CONFERENCE PROCEEDINGS - 1997 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-5     2533 - 2538  1997

     View Summary

    Bayesian theory is effective in statistics, lossless source coding, machine learning, etc. It is often, however, computationally expensive since the calculation of posterior probabilities and of mixture distributions is not tractable. In this paper, we propose a new method for approximately calculating mixture distributions in a discrete hypothesis class.

  • A new architecture of signature analyzers for multiple-output circuits

    TK Matsushima, T Matsushima, S Hirasawa

    SMC '97 CONFERENCE PROCEEDINGS - 1997 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-5     3900 - 3905  1997

     View Summary

    This paper presents a near architecture for multiple-input signature analyzers. The proposed signature analyzer with H delta inputs is designed by parallelizing a GLFSR(delta,m), where delta is the number of input signals and m is the number of stages in the feedback shift register. The GLFSR, developed by Pradhan and Gupta, is a general framework for rep resenting LFSR-based signature analyzers. The parallelization technique described in this paper can be applied to any kind of GLFSR signature analyzer, e.g., SISRs: MISRs, multiple MISRs and MLFSRs. It is show that a proposed signature analyzer with HS inputs requires less complex hardware than either single GLFSR(H delta,m)s or parallel construction H original GLFSR(delta,m)s. It is also shown that the proposed parallelization technique can be applied to a test pattern generator in BIST, since the GLFSR is also used to generate patterns for a CUT. The proposed technique would be practical for testing CUTs with a large number of input and output sequences, since the test circuit occupies a smaller area on the LSI chip than conventional test circuits.

  • Machine learning by a subset of hypotheses

    T Mukouchi, T Matsushima, S Hirasawa

    SMC '97 CONFERENCE PROCEEDINGS - 1997 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-5     2533 - 2538  1997

     View Summary

    Bayesian theory is effective in statistics, lossless source coding, machine learning, etc. It is often, however, computationally expensive since the calculation of posterior probabilities and of mixture distributions is not tractable. In this paper, we propose a new method for approximately calculating mixture distributions in a discrete hypothesis class.

  • A learning with membership queries to minimize prediction error

    Y Ukita, T Matsushima, S Hirasawa

    SMC '97 CONFERENCE PROCEEDINGS - 1997 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-5     4412 - 4417  1997

     View Summary

    In this paper, we consider the problem to predict the class of an unknown sample after learning from queries, We propose to evaluate a learning algorithm by a loss function for the prediction under a constraint, In this paper, the error probability for the prediction and the number of queries is defined as the loss function and the constraint, respectively, Then our objective is to minimize the error probability, the error probability is determined by what presentation order for instances to query and how to predict, Since the optimal prediction has been shown in previous researches, we only have to select the optimal presentation order for instances to query, We propose a lower bound used in the branch-and-bound algorithm to select the optimal presentation order for instances, Lastly, we show the efficiency of the algorithm using the derived lower bound by numerical computation.

  • Asymptotic property of sufficient statistic codes

    Toshiyasu Matsushima, Shigeichi Hirasawa

    IEEE International Symposium on Information Theory - Proceedings     421  1997

     View Summary

    In this paper, we obtain an accurate asymptotic formula of the Bayes optimal codelength in the case that the sources are not always i.i.d. sources by deriving the asymptotic codelength of sufficient statistic codes. © 1997 IEEE.

    DOI

  • The 55th Technical Comference(<Features>Frontiers of QC Activities in Production Fields)

    Toshiyasu Matsushima

      vol.27 ( no.4 ) 65 - 78  1997

  • Based on Bayes Statistics the On-line Learning Model and its Learnability

    Makoto Nakazawa, Toshiyasu Matsushima, Shigeichi Hirasawa

        519 - 520  1997

  • Document retrieval models using keywords

    Tsuneji Futagami, Toshiyasu Matsushima, Shigeichi Hirasawa

        257 - 258  1997

  • Parallel encoder and decoder architecture for cyclic codes

    TK Matsushima, T Matsushima, S Hirasawa

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E79A ( 9 ) 1313 - 1323  1996.09  [Refereed]

     View Summary

    Recently, the high-speed data transmission techniques that have been developed for communication systems have in turn necessitated the implementation of high-speed error correction circuits. Parallel processing has been found to be an effective method of speeding up operations, since the maximum achievable clock frequency is generally bounded by the physical constraints of the circuit. This paper presents a parallel encoder and decoder architecture which can be applied to both binary and nonbinary cyclic codes. The architecture allows H symbols to be processed in parallel, where H is an arbitrary integer, although its hardware complexity is not proportional to the number of parallel symbols H. As an example, we investigate hardware complexity for a Reed-Solomon code and a binary BCH code. It is shown that both the hardware complexity and the delay for a parallel circuit is much less than that with the parallel operation of H conventional circuits. Although the only problem with this parallel architecture is that the encoder's critical path length increases with H, the proposed architecture is more efficient than a setup using H conventional circuits for high data rate applications. It is also suggested that a parallel Reed-Solomon encoder and decoder, which can keep up with optical transmission rates, i.e., several giga bits/sec, could be implemented on one LSI chip using current CMOS technology.

  • 統計モデル選択の概要

    オペレーションズ・リサーチ   41 ( 7 ) 369  1996

  • A two-stage universal coding procedure using sufficient statistics

    T Matsushima, S Hirasawa

    PROCEEDINGS 1995 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY     17 - 17  1995

  • A two-stage universal coding procedure using sufficient statistics

    T Matsushima, S Hirasawa

    PROCEEDINGS 1995 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY     17 - 17  1995

  • A Bayes coding algorithm for FSM sources

    T Matsushima, S Hirasawa

    PROCEEDINGS 1995 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY     388 - 388  1995

  • 知識情報処理の基礎とその応用研究部会

    Shigeichi Hirasawa, Toshiyasu Matsushima

      vol.5 ( no.2 ) 140 - 142  1995

  • On Knowledge Representation and Reasoning System with a Range of Certainty Factors

    SUZUKI Makoto, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      35 ( 5 ) 691 - 705  1994.05

     View Summary

    We discuss the method for representing uncertain knowledge information and also on a reasoning system that guarantees a given reliability based on information theory and probability theory. In this paper, we propose a new reasoning system which consists of inductive and deductive inferences, by using a knowledge representation method with a range of certainty factors. In inductive inference, a new atom is computed from the observed data or facts, together with the upperbound and lowerbound between which its probability must fall. Also in deductive inference, which includes negation, conjunction, disjunction, chain and modus ponens, a new atom is deduced with a range of certainty factors computed from the results of inductive inference. Properties and behaviors of the system are clarified. It can be concluded that the new reasoning system is effective of conventional expert systems.

    CiNii

  • On block coded ARQ schemes cumulating the rejected sequences

    Toshihiro Niinomi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEEE International Symposium on Information Theory - Proceedings     376  1994  [Refereed]

     View Summary

    We show that the received sequences, which are rejected and not discarded on ARQ schemes, make the next decision with repeat request powerful by a random coding arguments. We first give the ordinary ARQ of the basic algorithm with some criterion, and define the ARQ with cumulating the rejected sequences, which naturally extended from the ordinary one. Then we study the error exponent of decoding the second received sequences under the condition that the first is rejected. We assume the block coded ARQ with noiseless feedback. © 1994 IEEE.

    DOI

  • A Bayes coding algorithm using context tree

    Toshiyasu Matsushima, Shigeichi Hirasawa

    IEEE International Symposium on Information Theory - Proceedings     386  1994  [Refereed]

     View Summary

    The context tree weighting (CTW) algorithm [Willems et al., 1993] has high compressibility for universal coding with respect to FSMX sources. The present authors propose an algorithm by reinterpreting the CTW algorithm from the viewpoint of Bayes coding. This algorithm can be applied to a wide class of prior distribution for finite alphabet FSMX sources. The algorithm is regarded as both a generalized version of the CTW procedure and a practical algorithm using a context tree of the adaptive Bayes coding which has been studied in Mataushima et al. (1991). Moreover, the proposed algorithm is free from underflow which frequently occurs in the CTW procedure. © 1994 IEEE.

    DOI

  • AN INDUCTIVE INFERENCE PROCEDURE TO MINIMIZE PREDICTION ERROR

    T MATSUSHIMA, H INAZUMI, S HIRASAWA

    IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS   23 ( 2 ) 547 - 554  1993.03  [Refereed]

     View Summary

    Considering inductive inference and deductive inference as not individual processes but a serial process of information processing, the serial inference procedure is studied from two purposes: one is the compression of observed facts, and the other is the prediction of new wffs. A new scheme of the serial inference process is proposed by using the correspondence to source coding and prediction problem. The optimal inference procedures for the two purposes are shown in the proposed scheme.

  • Learning formal languages from Feasible Teachers

    Tetsuya Sakai, Shigeichi Hirasawa, Toshiyasu Matsushima

      vol.4 ( no.3 ) 245 - 245  1993

  • 信頼性を考慮した推論について

    Makoto Suzuki, Toshihiro Niinomi, Shigeichi Hirasawa, Toshiyasu Matsushima

      vol.44 ( no.3 ) 245 - 245  1993

  • Inductive and Deductive Inference from the View Point of Information

    Toshiyasu Matsushima

    Memoirs of the school of Sci.& Eng. Waseda Univ.   vo.57   237 - 254  1993

  • A Model for Learning from a View Point of Information Theory

    Production Research 1993   17,385  1993

  • 不確実性をもつ論理式の帰納推論に関する一考察

    情報処理学会論文誌   33 ( 12 ) 1461 - 1475  1992.12

  • 帰納推論の情報理論による定式化と最適化

    横浜商大論集   25 ( 1-2 ) 120 - 188  1992

  • MDLの帰納推論への応用

    人工知能学会誌   7 ( 4 ) 615 - 621  1992

  • A CLASS OF DISTORTIONLESS CODES DESIGNED BY BAYES DECISION-THEORY

    T MATSUSHIMA, H INAZUMI, S HIRASAWA

    IEEE TRANSACTIONS ON INFORMATION THEORY   37 ( 5 ) 1288 - 1293  1991.09  [Refereed]

     View Summary

    The problem of distortionless encoding when the parameters of the probabilistic model of a source are unknown is considered from a statistical decision theory point of view. A class of predictive and nonpredictive codes is proposed that are optimal within this framework. Specifically, it is shown that the codeword length of the proposed predictive code coincides with that of the proposed nonpredictive code for any source sequence. A bound for the redundancy for universal coding is given in terms of the supremum of the Bayes risk. If this supremum exists, then there exists a minimax code whose mean code length approaches it in the proposed class of codes, and the minimax code is given by the Bayes solution relative to the prior distribution of the source parameters that maximizes the Bayes risk.

  • 不確実性をもつ仮説に関する帰納推論

    Toshiyasu Matsushima, Hiroshige Inazumi, Shigeichi Hirasawa

      vol.42 ( no.3 ) 212 - 212  1991

  • A CLASS OF NOISELESS CODES DESIGNED BY DECISION THEORY

    Toshiyasu Matsushima

      vol.7   381 - 405  1991

  • 情報理論による不確実な知識の表現法と推論に関する一考察

    早稲田大学理工学研究所報告   131  1991

  • An efficient user interface based on maximizing shared information

    Joe Suzuki, Toshiyasu Matsushima, Shigeichi Hirsawa, Hiroshige Inazumi

    Electronics and Communications in Japan (Part III: Fundamental Electronic Science)   73 ( 5 ) 40 - 49  1990  [Refereed]

     View Summary

    In designing systems with a human‐computer interface with a minimum number of interactions, there are two issues to consider: the determination of a proper sequence of questions by the user, and proper termination by the computer system, based on previous instructions. In short, these issues are those of feature selection ordering and a stopping rule for pattern recognition processes. Conventional treatments of these problems have been investigated from the viewpoint of an average error rate. When the number of patterns is large, however, and the number of instructions to be terminated is large, from the point of view of user interface efficiency, and the average error rate is not an effective indicator at the intermediate stage. In this paper, a new method is proposed for determining feature selection ordering and a stopping rule which focuses on the remaining patterns at each stage, and which maximizes the value of mutual information between the user's responses and the required pattern. Important properties associated with this scheme have been demonstrated while evaluating its performance via computer simulation. One is that the average information gain at each stage decreases monotonically, and another is that this scheme produces the minimum error rate. Copyright © 1990 Wiley Periodicals, Inc., A Wiley Company

    DOI

  • Inductive Inference scheme at a Finite Stage of Process from View Point of Source Coding

    信学論(E)   73E ( 5 ) 644 - 652  1990

  • Stopping rules based on a posteriori probability for every pattern and every stage

    Joe Suzuki, Toshiyasu Matsushima, Shigeichi Hirasawa, Hiroshige Inazumi

    Electronics and Communications in Japan (Part III: Fundamental Electronic Science)   72 ( 8 ) 71 - 82  1989  [Refereed]

    DOI

  • 相互情報量最大に基準を置くユーザインタフェースの効率化

    Joe Suzuki, Toshiyasu Matsushima, Hiroshige Inazumi, Shigeichi Hirasawa

      vol.J72-A ( no.3 ) 517 - 524  1989  [Refereed]

  • 情報理論に基づく推論の体系化と不確実な知識表現への応用

    Toshiyasu Matsushima, Joe Suzuki, Hiroshige Inazumi, Shigeichi Hirasawa

      vol.40 ( no.3 ) 196 - 196  1989

  • Applications of Information Theory into Predicate Logic for AI Systems

    IEICE TRANSACTIONS   Vol.E72-E ( No.5 ) 443 - 451  1989

  • パターンごと・ステージごとに事後確率のしきい値をおくストッピングルール

    Joe Suzuki, Toshiyasu Matsushima, Hiroshige Inazumi, Shigeichi Hirasawa

      vol.J71-A ( no.6 ) 1299 - 1308  1988  [Refereed]

  • 質問応答回数最小をねらいとした効率的な知的インタフェイスの設計

    Joe Suzuki, Toshiyasu Matsushima, Hiroshige Inazumi, Shigeichi Hirasawa

      vol.39 ( no.3 ) 191 - 191  1988

  • あいまいな命題を含む推論モデルに関する一考察

    Toshiyasu Matsushima

      vol.37 ( no.5 ) 326 - 326  1986

  • Variable-Length Intrinsic Randomness Allowing Positive Value of the Average Variational Distance

    Yoshizawa, Jun, Saito, Shota, Matsushima, Toshiyasu

    Proceedings of 2018 International Symposium on Information Theory and its Applications, ISITA2018  

  • Linear Programming Bounds for multi-level Unequal Protection Codes

    Saito Tomohiko, Matsushima, Toshiyasu, Hirasawa, Shigeichi

    Proceedings of 2018 IEEE International Conference on Systems, Man, and Cybernetics  

  • New Results on Variable-Length Lossy Compression Allowing Positive Overflow and Excess Distortion Probabilities

    Saito, Shota, Yagi, Hideki, Matsushima, Toshiyasu

    Proceedings of 2018 International Symposium on Information Theory and its Applications, ISITA2018  

  • Sparse Bayesian Hierarchical Mixture of Experts and Variational Inference

    Iikubo, Yuji, Horii, Shunsuke, Matsushima, Toshiyasu

    Proceedings of 2018 International Symposium on Information Theory and its Applications, ISITA2018  

  • Bayesian Independent Component Analysis under Hierarchical Model on Independent Components

    Asaba, Kai, Saito, Shota, Horii, Shunsuke, Matsushima, Toshiyasu

    Proceedings of 2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2018  

  • Cumulant Generating Function of Codeword Lengths in Variable-Length Lossy Compression Allowing Positive Excess Distortion Probability

    Saito, Shota, Matsushima, Toshiyasu

    Proceedings of 2018 International Symposium on Information Theory, ISIT2018  

  • A Note on Weight Distributions of Spatially "Mt. Fuji" Coupled LDPC Codes.

    Nakahara, Yuta, Matsushima, Toshiyasu

    IEICE Transactions   vol. 101-A ( 12 )

  • Variable-Length Intrinsic Randomness on Two Performance Criteria based on Variational Distance

    Jun Yoshizawa, Shota Saito, Toshiyasu Matsushima

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E102-A ( 12 )  [Refereed]

  • Asymptotic Property of Universal Lossless Coding for Independent Piecewise Identically Distributed Sources

      - 1

  • Variable-Length Lossy Compression Allowing Positive Overflow and Excess Distortion Probabilities

    ISIT2017  

  • Parallel Concatenation of Polar Codes and Iterative Decoding

     

  • Asymptotics of MLE - based Prediction for Semi - supervised learning

     

  • On the Bayesian Forecasting Algorithm under the Non-Stationary Binomial Distribution with the Hyper Parameter Estimation

     

  • A Formalization of Generalized Probabilistic Reasoning and Its Procedure on Junction Trees

     

  • Bayesian Forecasting of WWW Traffic on the Time Varying Poisson Model

     

  • On the Hyperparameter Estimation of Time Varying Poisson Model for Bayesian WWW Traffic Forecasting

     

  • Multiuser Detection Algorithm for CDMA based on the Belief Propagation Algorithm

     

  • A Note on ANOVA in an Experimental Design Model Based on an Orthonormal System

    2012 IEEE International Conference on Systems, Man, and Cybernetics (SMC 2012)  

  • On the evaluation of the achievable codelength of Fixed-length codes

     

  • Robustness of Syndrome Analysis Method in Highly Structured Fault-Diagnosis Systems

    2014 IEEE International Conference on Systems, Man and Cybernetics - SMC  

  • A Note on Relation Between the Fourier Coefficients and the Interaction Effects in the Experimental Design

     

  • A Note on Relation between the Fourier Coefficients and the Effects in the Experimental Design

     

  • A Note on Linear Programming Based Communication Receivers

     

  • On Uncertain Logic Based upon Information Theory

     

  • On the Optimal Inductive Inference Scheme from the View Point of Source Coding

     

  • A Bayes coding algorithm for FSMX soureces

     

  • An analysis on Difference between the Code based on MDL Principle and the Bayes Code(共著)

     

  • A Note on Concept Learning with Membership Queries(共著)

     

  • Asymptotic Property of Sufficient Statistic Codes

     

  • A Study on Difference of Codelength between MDL Codes and Bayes Codes in Case Different Prior are Assumed(共著)

     

  • A Learning with Membership Queries to Minimize Prediction Error(共著)

     

  • A study on difference of codelenghs between MDL codes and Bayes codes on case different priors are assumed

     

  • On a Reasoning Model with a Knowledge Representation Matrix for Uncertainty

     

  • Non-Hierarchical Clustering using Genetic Algorithm(GA) and K-means Method

     

  • Iterative Soft-Decision Decoding Using an Update Algorithm of Error-Locator Polynomial for Binary BCH Codes

     

  • Consistency of Bayesian Model Selection

     

  • Compression of File for Document Retrieval System

     

  • An Approximation Algorithm of Bayes Coding for FSMX sources

     

  • A Universal Code Considering the Codeword Cost

     

  • A Note on Learning of Probabilistic Hypotheses from Membership Queries

     

  • On a Reasoning Model with a Knowledge Representation Matrix for Uncertainty

     

  • Non-Hierarchical Clustering using Genetic Algorithm (GA) and K-means Method

     

  • Iterative Soft-Decision Decoding Using an Update Algorithm of Error-Locator Polynomial for Binary BCH Codes

     

  • Consistency of Bayesian Model Selection

     

  • Compression of File for Document Retrieval System

     

  • An Approximation Algorithm of Bayes Coding for FSMX sources

     

  • A Universal Code Considering the Codeword Cost

     

  • A Note on Learning of Probabilistic Hypotheses from Membership Queries

     

  • On a Deductive Reasoning Model and Method for Uncertainty

     

  • An interpretation of Turbo decoding from the viewpoint of differential geometry

     

  • An Iterative Algorithm for Calculating Posterior Probability and Model Representation

     

  • On the evaluation of the achievable codelength of Fixed-length codes

     

  • A Note on Spelling Correction Methods based upon Statistical Decision Theory

     

  • An Information Spectrum Consideration on the Universal Joint Source-Channel Coding

     

  • A Note on the Branch-and-Cut Approach to Decoding Linear Block Codes

     

  • A Description of Experimental Design using an Orthonormal System

     

  • Linear Programming Bounds of Orthogonal Arrays for Experimental Designs

     

  • Bayes universal source coding scheme for correlated sources

     

▼display all

Misc

  • A Note on Optimality of Orthogonal Designs in Complex Space

    Yoshifumi Ukita, Toshiyasu Matsushima

      1   171 - 172  2020.03

  • A Study on Simultaneous Experiments for Related Linear Models based on an Orthonormal System

    Yoshifumi Ukita, Toshiyasu Matsushima

    Bayes on the Beach 2019    2019.11  [Refereed]

  • On decoding of Polar codes using search algorithm

      115 ( 214 ) 55 - 60  2015.09

    CiNii

  • Special Section on Information Theory and Its Applications FOREWORD

    Toshiyasu Matsushima

    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES   E97A ( 12 ) 2287 - 2287  2014.12

    Other  

  • A Note on Attack Against Nonlinear Combiner Generator Using Sum-Product Algorithm

    KUBO Kota, SAITO Shota, KAMATSUKA Akira, MATSUSHIMA Toshiyasu

      114 ( 306 ) 357 - 364  2014.11

     View Summary

    When parameters have some restrictions, we can present it on a graphical model and find the values of unknown parameters from the parameters that we know. The one method to estimate the values of unknown parameters is to apply Sum-Product Algorithm(SPA). This technique is also discussed in cryptology. In this paper, finding the value of unknown parameters in stream cipher by using SPA is discussed. Some loops exist in the graph, thus the result of SPA is not guaranteed. Therefore, we evaluate calculating posterior probabilities using SPA by changing shape of the graphical model, and by changing message passing schedule.

    CiNii

  • How to Observe Probability Distribution of Self-Information as viewed under different resolutions of n^<-1> and n^<-1/2>

    MATSUSHIMA Toshiyasu

    IEICE technical report. Information theory   114 ( 138 ) 83 - 88  2014.07

     View Summary

    Several problems in Information Theory are investigated by asymptotic approximations of the probability of self-information or self-mutual information. Two kind of normal asymptotic approximation is discussed in this paper. One is the approximation of tail distribution, and the other is the approximation of top distribution that is the neighborhood of the mode in the distribution. Error probability and overflow probability are evaluated by the approximation of tail distribution. Mean code length and point wise code word length are evaluated by the approximation of top distribution.

    CiNii

  • A study of a MDP-based recommender system when the class of user who is recommended items is unknown

    IWAI Shusuke, MIYA Nozomi, MAEDA Yasunari, MATSUSHIMA Toshiyasu

    IEICE technical report. Information theory   114 ( 138 ) 49 - 54  2014.07

     View Summary

    Recommender system is a system that gives users objects, information and items they want. In previous studies, recommendation was once and result of recommendation was not considered. Recently, in contrast, studies that consider results of recommendation have done. For example, there are studies that apply marcov decision process to recommender system. In addition, there are studies that apply marcov decision process and statistical decision theory to recommender system. These studies consider all users' probability of purchase are same. In contrast, we consider users are belongs to a class that classified by similarity. In addition, we consider the probability of purchase by users belongs to the class is different from the probability of purchase by users belongs to other class. We consider classes of users whose purchase data we have are known and class of the user who is recommended item is unknown. Then, we use marcov decision process and statistical decision theory. We renovate decision function by maximizing bayes criterion when the user who is recommended items buy new item.

    CiNii

  • A Note on Optimal Control System for Selective-Repeat Hybrid SR-ARQ with a Finite Length Buffer

    KAGEYAMA Yuta, KAMATSUKA Akira, MAEDA Yasunari, MATSUSHIMA Toshiyasu

    IEICE technical report. Information theory   114 ( 138 ) 55 - 58  2014.07

     View Summary

    Hybrid ARQ is one of the error correction scheme. It is Combination of ARQ and FEC. Algorithms for throughput maximization were proposed. But their efficiency was not guaranteed in theory. In this study, A way of acknowledgement decision is formulated by Markov decision process in terms of statical decision theory. The algorithm of throughput maximization which is considered reliablity is derived.

    CiNii

  • Iterative Multiuser Joint Decoding based on Augmented Lagrangian Method

    HORII Shunsuke, SUKO Tota, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    IEICE technical report. Information theory   113 ( 228 ) 13 - 17  2013.09

     View Summary

    In this paper, we develop an iterative multiuser joint decoding of code-division multiple-access (CDMA) signals based on a distributed optimization algorithm. For the joint decoding problem, decoding algorithm based on the turbo principle is widely used. The algorithm consists of soft-input soft-output (SISO) channel decoder and SISO multiuser detector and it can be derived as an application of the sum-product algorithm. On the other hand, in the research area of error correcting codes, the decoding algorithm based on convex optimization has been attracting a great deal of attention. Decoding algorithm based on linear programming (LP) has decoding error rate which is comparable with sum-product algorithm with stronger theoretical guarantees. We formulate the joint decoding problem of CDMA signals as an convex optimization problem and we present a relax form of the problem. Moreover, we propose a distributed algorithm which efficiently solves the relaxed optimization problem. The proposed algorithm is based on alternating direction method of multipliers (ADMM).We also see the performance of the proposed decoder through numerical simulations.

    CiNii

  • A Consideration on Modeling and Optimality of Regenerating Codes Considering Security of Shares

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    IEICE technical report. Information theory   113 ( 153 ) 27 - 32  2013.07

     View Summary

    We consider regenerating codes. Regenerating codes are a class of codes for distributed storage systems that enable a data collector to recover the original data by connecting to any k of n storage nodes, and also can repair a failed node by downloading data from any d (≧=k) nodes. In regenerating codes, there exists a tradeoff between the storage size of each node and repair-bandwidth. In this study, we define two classes of regenerating codes considering security of each node's share, and show that the regenerating codes, which was proposed by Rashmi, Shah and Kumar, achieve the minimum storage size and repair-bandwidth for our class of regenerating codes.

    CiNii

  • A Consideration on Minimum Storage Regenerating Codes for Functions

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    Technical report of IEICE. ISEC   112 ( 461 ) 107 - 112  2013.03

     View Summary

    We consider regenerating codes for functions. Regenerating codes are a class of codes for distributed storage systems that enable a data collector to recover the original data by connecting to any k of n storage nodes, and also can repair a failed node by downloading data from any d(≧k)nodes. In this study, we present explicit construction of regenerating codes for function (p(.)that enable a data collector to compute (p(x)for input J by connecting to any k;nodes and sending J to all connected nodes. The presented codes can repair a failed node by downloading data from any d (≧k)nodes as with regenerating codes, and achieve the minimum storage capacity.

    CiNii

  • Hierarchical Multi-label Classification on Statistical Decision Theory

    YAMAMOTO Kiyohito, SUKO Tota, MATSUSHIMA Toshiyasu

      112 ( 452 ) 101 - 106  2013.03

     View Summary

    This paper considers multi-label classification on statistical decision theory. In Label Power Set format, multi-label classification is equivalent to multi-class classification. However, the number of classes increases exponentially as elements in label set grow in number. Hence in case of many labels, a prohibitive computational cost problem occurs. To avoid this problem, some studies have been done and one of them used hierarchical structure. On the other hand, optimal classification method based on bayes rule has been attracted much attention recently. We apply this optimal classification method based on bayes rule to multi-label classification problem. Moreover, assuming hierarchical structure on labels, we propose efficient classification algorithms which reduce computational cost to linear order on the number of elements in label set. Since optimal classification based on bayes rule differs calculation formula depending loss function, we present algorithms in case of O-1 loss and hamming loss, respectively.

    CiNii

  • A Privacy Preserving Distributed Calculation Method of Least-squares Estimator for Linear Regression Models

    SUKO Tota, HORII Shunsuke, KOBAYASHI Manabu, GOTO Masayuki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      112 ( 279 ) 107 - 111  2012.10

     View Summary

    In this paper, we study a privacy preserving linear regression analysis. We propose a new protocol of distributed calculation method that calculates a least squares estimator, in case that two parties have different types of explanatory variables. We show security of privacy in the proposed protocol. Because the protocol have iterative calculation, we evaluate the number of iterations with numerical experiments. Finally, we show an extended protocol that is a distributed calculation method for κ parties.

    CiNii

  • Linear Programming Decoding for Multiple Access Channel based on Decomposition Methods

    HORII Shunsuke, MATSUSHIMA Toshiyasu

    IEICE technical report. Information theory   112 ( 215 ) 53 - 58  2012.09

     View Summary

    In this paper, we develop a dual decomposition method based linear-programming (LP) decoding for multiple-access channels with binary linear codes. LP decoding has decoding error rate which is comparable with state-of-the-art Sum-Product decoder for the decoding problem of binary linear codes with stronger theoretical guarantees and it is applied to decoding problems over various channels such as interference channels and multiple-access channels. But the decoder has to solve an LP problem and in general its computational complexity is the polynomial order of the number of variables and the number of constraints. Therefore it needs more decoding time compared to other decoders such as Sum-Product decoder. In order to tackle the complexity problem, many complexity efficient decoders have been developed for single-user memoryless channels. Especially, some previous studies showed that the decoding algorithm based on dual decomposition works well for large scale LDPC codes. In this paper, we apply the dual decomposition methods for the decoding problem over multiple access channels.

    CiNii

  • マルコフモデルによる自動分類に対する分類誤り確率の推定

    小林学, 後藤正幸, 松嶋敏泰, 平澤茂一

    全国大会講演論文集   2012 ( 1 ) 61 - 63  2012.03

     View Summary

    カテゴリが既知の学習データを用いて,新規データのカテゴリを推定する自動分類問題は,サポートベクターマシンなど数多く研究が行われている.本研究では,カテゴリ中に発生するデータがマルコフモデルに従って発生すると仮定したときに,学習データを用いて分類誤り確率を推定する手法を提案する.

    CiNii

  • 実験計画法に適した直交配列の線形計画限界

    斉藤 友彦, 浮田 善文, 松嶋 敏泰, 平澤 茂一

    情報処理学会第74回全国大会     261 - 262  2012.03

  • An Error Probability Analysis of the Text Classification Using the CTW Algorithm

    KOBAYASHI Manabu, GOTO Masayuki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    IEICE technical report. Nonlinear problems   111 ( 276 ) 109 - 114  2011.11

     View Summary

    The text classification problem has been investigated by various techniques, such as a vector space model, a support vector machine, and so on. On the other hand, the Context-Tree Weighting (CTW) algorithm has been proposed as an outstanding data compression. Furthermore, experimental results have been reported using the CTW algorithm for the text classification. In this paper, we assume that each document with same category arises from one stochastic model for the text classification using the CTW algorithm. Then we propose an analysis method to obtain the classification error probability for the document with the finite length.

    CiNii

  • 実験計画法における効果の推定の計算量削減に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    第31回情報理論とその応用シンポジウム予稿集     945 - 950  2008.10

  • 直交計画と双対符号の関係に関する一考察

    浮田 善文, 松嶋 敏泰

    電子情報通信学会研究技術報告   IT2006-107   131 - 136  2007.03

  • 語頭条件を満たさないWord-valued sourceに対するLZ78符号の符号化性能について

    石田崇, 松嶋敏泰, 平澤茂一

    電子情報通信学会, 技術研究報告, vol. 106, no. 516, IT2006-52, pp. 13-18, 2007年1月.    2007

  • Shortened and Concatenated Collusion-secure Codes for Digital Fingerprinting

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Technical Report, vol. 107,no.42, IT2007-6, pp.31-36, May 2007.    2007

  • An Adaptive Decoding Algorithm of LDPC Codes over the Binary Erasure Channel

    Gou Hosoya, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. 2007 Hawaii and SITA Joint Conference on Information Theory, Hawaii, U.S.A., May 2007.    2007

  • Shortening Methods of Collusion-Secure Codes for Digital Fingerprinting

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. 2007 Hawaii and SITA Joint Conference on Information Theory, Hawaii, U.S.A., May 2007.    2007

  • A Generalization of the Parallel Error Correcting Codes by Allowing Some Random Errors

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    To be appeared at IEICE Trans. Fundamentals, vol.E90-A, no.10, Sep. 2007.    2007

    DOI

  • A Modification Method for Constructing Low-Density Parity-Check Codes for Burst Erasures

    Gou Hosoya, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Technical Report   IT2005-121   153 - 158  2006

  • On Factorial Effects Corresponding to Orthogonal Arrays with Unequal Strength

    Tomohiko Saito, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Technical Report   vol.106 ( no.60 ) 53 - 58  2006

  • A New Class of Traceability Codes for Digital Fingerprinting

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Technical Report   vol.106 ( no.60 ) 13 - 18  2006

  • New Traceability Codes against a Generalized Collusion Attack for Digital Fingerprinting

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proceedings of 2006 International Workshop on Information Security Applications (WISA2006) , Jeju Island, Korea, Aug. 2006.    2006

  • Fast Algorithm for Generating Candidate Codewords in Reliability-Based Maximum Likelihood Decoding

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Trans. Fundamentals,   E89-A ( 10 ) 2676 - 2683  2006

    DOI

  • A Modification Method for Constructing Low-Density Parity-Check Codes for Burst Erasures

    Gou Hosoya, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    IEICE Trans. Fundamentals,   E89-A ( 10 ) 2501 - 2509  2006

    DOI CiNii

  • Performance of Low-Density Parity-Check Codes for Burst Erasure Channels

    Gou Hosoya, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. of 2006 International Symposium on Information Theory and its Applications (ISTIA2006), Seoul, Korea, Oct. 2006.    2006

  • A Generalization of the Parallel Error Correcting Codes

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. of 2006 IEEE Information Theory Workshop (ITW2006), Chengdu, China, Oct. 2006.    2006

  • On Correctable Burst-Erasure Lengths for LDPC Codes with Column Permuted Parity-Check Matrices

    Gou Hosoya, Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. 29th Symposium on Information Theory and its Applications (STIA2006), pp.645-648, Hakodate, Nov. 2006.    2006

  • Decoding Performance of Linear Parallel Error Correcting Codes

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    Proc. 29th Symposium on Information Theory and its Applications (STIA2006), pp.189-192, Hakodate, Nov. 2006.    2006

  • HMM通信路に対するEM復号の復号誤り確率の評価法

    小林 学, 八木 秀樹, 松嶋 敏泰, 平澤 茂一

    第29回情報理論とその応用シンポジウム予稿集, 函館,2006年11月.    2006

  • A Note on the Construction of Orthogonal Designs Using Error Correcting Codes

    SAITO Tomohiko, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      27 ( 2 ) 463 - 466  2005.12

    CiNii

  • Redundancy of Bayes Codes for Nonstationary Sources with Piecewise Constant Parameters

    SUKO Tota, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      27 ( 2 ) 523 - 526  2005.12

    CiNii

  • A Note on the Asymptotics of Bayes Prediction for Hierarchical Models

    TAKUMI Takeo, SUKO Tota, MATSUSHIMA Toshiyasu

      27 ( 2 ) 639 - 642  2005.12

    CiNii

  • A Heuristic Search Algorithm with the Reduced List of Test Error Patterns for Maximum Likelihood Decoding

    YAGI Hideki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      27 ( 2 ) 571 - 574  2005.12

    CiNii

  • インターネットを用いた研究支援環境~電子会議システム~

    野村亮, 石田崇, 中澤真, 鴻巣敏之, 松嶋敏泰, 平澤茂一

    経営情報学会 2005年度春季全国研究発表大会予稿集     254 - 257  2005

  • インターネットトラヒックのポアソン分布の密度パラメータが時間変動する時系列モデルを用いた解析に関する一考察

    小泉大城, 松嶋敏泰, 平澤茂一

    2005年FIT(情報科学技術フォーラム)論文集   L-016  2005

  • LDPC符号の消失訂正と誤り訂正の関係

    細谷剛, 松嶋敏泰, 平澤茂一

    電子情報通信学会 技術研究報告   vol.105 ( no.311 ) 13 - 17  2005

  • アプリオリアルゴリズムを用いた指定精度を保証するマトリックスクラスタリング手法

    小林 学, 松嶋 敏泰, 平澤 茂一

    経営情報学会春季全国大会予稿集   vol.1   50 - 53  2005

  • 有限幾何に基づくFingerprintingのための結託耐性符号,

    八木秀樹, 松嶋敏泰, 平澤茂一

    第28回情報理論とその応用シンポジウム予稿集     701 - 704  2005

  • A Note on Improvement of a Fast Correlation Attack

    HOSOBUCHI Satoshi, SAITO Tomohiko, MATSUSHIMA Toshiyasu

      27 ( 1 ) 37 - 40  2004.12

    CiNii

  • Non-perfectly secure identity based decentralized key distribution system

    YOSHIDA Takahiro, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      27 ( 1 ) 327 - 330  2004.12

    CiNii

  • A Note on Decoding of Low Density Parity Check Codes with Small-Loop

    KOBAYASHI Naoto, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

      27 ( 1 ) 267 - 270  2004.12

    CiNii

  • 質問学習と逐次実験計画の関係に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    電子情報通信学会研究技術報告   AI2003-63   1 - 6  2004.01

  • Word-valued sourceから出現する系列の単語分割について

    石田崇, 松嶋敏泰, 平澤茂一

    第27回情報理論とその応用シンポジウム予稿集,岐阜     135 - 138  2004

  • A decoding method of low-dencity parity-check codes over binary symmetric channel

    Gou Hosoya, Toshiyasu Matsushima, Shigeichi Hirasawa

    第27回情報理論とその応用シンポジウム予稿集,岐阜    2004

  • "A Method for Reducing Space Complexity of Reliability based Heuristic Search Maximum Likelihood Decoding Algorithms" The 26th Symposium on Information Theory and Its Applications (SITA2003),

    Hideki Yagi, Toshiyasu Matsushima, Shigeichi Hirasawa

    第26回情報理論とその応用シンポジウム予稿集   pp.185-188  2003

  • "語頭条件を満たさないWord-Valued Sourceモデルに関する一考察"

    石田崇, 松嶋敏泰, 平澤茂一

    第26回情報理論とその応用シンポジウム予稿集   pp.77-80  2003

  • "語頭条件を満たさない単語集合をもつWord-Valued Sourceの性質について"

    石田崇, 後藤正幸, 松嶋敏泰, 平澤茂一

    電子情報通信学会技術報告   IT2003-5, pp23-28  2003

  • "インターネットを用いた研究支援システム"

    中澤真, 野村亮, 鴻巣敏之, 松嶋敏泰

    私立大学情報教育協会平成15年度大学情報化全国大会予稿集,東京   pp.72-73  2003

  • ブール関数の逐次実験計画を用いた学習に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    電子情報通信学会研究技術報告   COMP2002-52   47 - 53  2002.11

  • 単語単位で系列を出力する情報源の性質について

    石田崇, 後藤正幸, 松嶋敏泰, 平澤茂一

    第25回情報理論とその応用シンポジウム予稿集   pp.695-698  2002

  • 計算論的情報源モデルと圧縮アルゴリズム

    中澤真, 松嶋敏泰, 平澤茂一

    電子情報通信学会情報理論研究会   pp.19-24  2002

  • 「インターネットを用いたゼミと研究指導」実用化報告

    野村亮, 中澤真, 松嶋敏泰, 平澤茂一

    2002PCカンファレンス    2002

  • 計算論的学習と情報圧縮に関する一考察

    中澤真, 松嶋敏泰, 平澤茂一

    電子情報通信学会情報理論研究会   pp.37-42  2002

  • ベイズ決定理論による定式化のもとで直交計画を用いたブール関数の学習に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    電子情報通信学会研究技術報告   COMP2001-59   51 - 58  2001.11

  • ブロックターボ符号に対するインタリーバの構成法と最小距離

    小林学, 松嶋敏泰, 平澤茂一

    第24回情報理論とその応用シンポジウム   P.95~P.98  2001

  • 「インターネットを用いた研究活動支援システム」システム構成と評価

    野村亮, 中澤真, 鴻巣敏之, 松嶋敏泰, 平澤茂一

    2001年日本経営学会秋季研究発表大会    2001

  • 形式言語と圧縮に関する一考察

    中澤真, 松嶋敏泰, 平澤茂一

    電子情報通信学会情報理論研究会   P.19~P.24  2001

  • 「インターネットを用いた研究活動支援システム」システム構成

    平澤茂一, 松嶋敏泰, 鴻巣敏之, 酒井哲也, 中澤真, 李相協, 野村亮

    2001PCカンファレンス    2001

  • ブロックターボ符号の生成行列と性能評価

    小林学, 松嶋敏泰, 平澤茂一

    電子情報通信学会情報理論研究会   P.1~P.6  2001

  • フーリエ変換を用いたブール関数の学習に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    電子情報通信学会研究技術報告   COMP2000-56   49 - 55  2000.11

  • 不確実性を含む演繹推論に関する一考察

    鈴木誠, 松嶋敏泰, 平澤茂一

    第23回情報理論とその応用シンポジウム   (427-430)  2000

  • 遺伝的アルゴリズムとk-means法を用いたクラスタリングの効率化

    李相協, 松嶋敏泰, 平澤茂一

    日本経営工学会 春季発表大会   (120-121)  2000

  • A Study on Difference of Codelengths between Codes based on MDL Principle and Bayes Codes for Given Prior Distributions

    Masayuki Goto, Toshiyasu Matsushima, Shigeichi Hirasawa

    Electronics and Communications in Japan, Part 3   84,4(30-40)  2000

  • 不確実な知識の演繹推論アルゴリズムに関する一考察

    小島 健司, 浮田 善文, 松嶋 敏泰

    第22回情報理論とその応用シンポジウム予稿集     841 - 844  1999.12

  • 質問からのブール関数の学習における学習戦略を求めるアルゴリズム

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    第22回情報理論とその応用シンポジウム予稿集     845 - 848  1999.12

  • 音声認識における Hidden Markov Model のパラメータ推定に関する一考察

    高井 望, 浮田 善文, 松嶋 敏泰

    第21回情報理論とその応用シンポジウム予稿集     643 - 646  1998.12

  • 不確実な知識の推論における欠測データの取り扱いに関する一考察

    金澤 裕輔, 浮田 善文, 松嶋 敏泰

    第21回情報理論とその応用シンポジウム予稿集     153 - 156  1998.12

  • 適合フィードバックによる情報検索に関する一考察

    鍛治 美緒, 浮田 善文, 松嶋 敏泰

    第21回情報理論とその応用シンポジウム予稿集     161 - 164  1998.12

  • 対数線形モデルを用いた不確実な知識の推論法について

    小島 健司, 浮田 善文, 松嶋 敏泰

    人工知能学会全国大会(第12回)論文集     308 - 311  1998.06

  • マルコフ決定過程の計算アルゴリズムについて

    井村 真樹, 浮田 善文, 松嶋 敏泰

    人工知能学会全国大会(第12回)論文集     56 - 59  1998.06

  • 決定を考慮したベクトル量子化法の提案

    高井 望, 浮田 善文, 松嶋 敏泰

    第20回情報理論とその応用シンポジウム予稿集     609 - 612  1997.12

  • 直交表現された仮説の学習に関する一考察

    峯松 一彦, 浮田 善文, 松嶋 敏泰

    第20回情報理論とその応用シンポジウム予稿集     701 - 704  1997.12

  • 質問からの学習における予測誤りに関する一考察

    浮田 善文, 松嶋 敏泰

    第20回情報理論とその応用シンポジウム予稿集     697 - 700  1997.12

  • 未知パラメータを含むマルコフ決定過程における学習アルゴリズムの提案

    前田 康成, 浮田 善文, 松嶋 敏泰, 平澤 茂一

    第19回情報理論とその応用シンポジウム予稿集     597 - 600  1996.12

  • 不確実性を持つ概念の質問による学習に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    人工知能学会全国大会(第10回)     75 - 78  1996.06

  • Parallel Architecture for Signature Analyzer in LSI Self-Testing

    MATSUSHIMA Tomoko K., MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    Technical report of IEICE. ICD   96 ( 21 ) 85 - 92  1996.04

     View Summary

    Several kinds of LFSR-based signature analyzers have been proposed for LSI built-in self test. MLFSR, which has been proposed for multiple output CUT by Pradhan et al., has the possibility to achieve the better aliasing probability than other schemes such as the multiple MISR scheme with comparable hardware complexity. The only problem with this scheme is that the hardware complexity becomes larger than propotionally as δ increases, where δ is the number of parallel signals inputted into MLFSR, and so MLFSR is not adequate for testing CUTs with large number of output signals. In this report, we propose a parallel architecture for signature circuits applicable to such CUTs. This architecture allows Hδ bits to be processed in parallel, where H can be chosen as an arbitrary integer. It is shown that the complexity for the parallel circuit, with Hδ input, signals is much smaller than that for a parallel opereation of H converntional circuits with δinput signals.

    CiNii

  • 構成的帰納論理プログラムに関する一考察

    新津 健, 浮田 善文, 松嶋 敏泰

    電子情報通信学会技術研究報告   AI95-47   41 - 46  1996.01

  • A Study on Parallel Encoder and Decoder for Cyclic Codes

    MATSUSHIMA Tomoko K., MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    IEICE technical report. Information theory   95 ( 347 ) 23 - 28  1995.11

     View Summary

    In this paper, a parallel encoder and decoder architecture for cyclic codes is presented. This architecture can decrease encoding and decoding time 1/H(H > 1) compared to conventional codec, where H is the number of symbols processed in parallel. As an example, we investigate hardware complexity for a (255,251) Reed-Solomon code over GF(2^8). It is shown that hardware complexity for a set of parallel encoder and decoder is much smaller than that for a parallel operation of H conventional sets.

    CiNii

  • 質問を許す概念学習に関する一考察

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    第18回情報理論とその応用シンポジウム予稿集     537 - 540  1995.10

  • PROLOGを対象とした帰納的学習の効率化

    浮田 善文, 松嶋 敏泰, 平澤 茂一

    人工知能学会全国大会(第8回)     137 - 140  1994.06

▼display all

Research Projects

  • ビジネス価値創造のためのデータ解析プラットフォームと時変協調フィルタリングの研究

    Project Year :

    2019.04
    -
    2023.03
     

     View Summary

    企業の持つビッグデータを有効に活用して新たなビジネス価値創造を行うことは,日本において急務である.またこれを産学連携で行うには,個人情報やセキュリティに配慮した上でデータ解析を行うための効果的なプラットフォームと,その運用方法の確立が重要である.本研究ではデータを持ち出さずに解析を行うプラットフォームの構築を行い,その効果的な運用方法の設計を行う.さらにライフイベントなど時間で変化する状況における顧客の消費行動を数理モデル化し,企業データを活用することによりビジネス価値創造へつなげる手法の提案を行う

  • AIコーチによるプログラミング独習システム

    Project Year :

    2018.04
    -
    2021.03
     

     View Summary

    本研究は「回答履歴から学習者の弱点を推定し適切な問題を推薦するプログラミング演習Webシステム」の構築を目的とする.本システムの学術的独創性・創造性は次の二点にある.(i)学習者の演習回答時における編集履歴と脳波からその演習に対する理解度を推定する.(ii)オンライン・ショッピングで用いられる推薦システムを使い問題を推薦する.特に学習者への推薦・回答履歴とその他大勢の履歴を使った時空間推薦アルゴリズムを使う.研究期間内における本研究の最終目標は,システムの完成及び評価である.具体的手順は「Step 1. 基礎理論の構築:学習者の理解度推定法,推薦アルゴリズムを完成させる」,「Step 2. 実装:システムの実装を行う」,「Step 3. 実験:高校生と大学生に本システムを利用してもらい,検証を行う」,「Step 4. 評価:Step 3. の結果に基づき,本システムを評価する」である.本年度は,「Step 2. 実装:システムの実装を行う」を主に行った.本年度の主な成果はいくつかのオープンソースを参考にオンラインジャッジサーバーを実際に構築したことである.これによって,これまで積み上げた基礎理論の実装が可能となった.また,昨年度までに構築した,ポートフォリオ,学習管理システム,オンラインコンパイラと組み合わせ実用的なシステムの構築と有効性に関する検証実験が可能となった.現在は実際にいくつかのプログラミング演習問題を作成・掲載し,小規模ではあるが運用を行っている.これらの予備実験を重ね,実用的な運用に耐えられるかの検証を行っている.本年度の主な研究成果はオンラインジャッジサーバの構築が完成したことである.昨年度までの研究成果であるポートフォリオ,学習管理システム,オンラインコンパイラと合わせ,サーバの構築は順調に進んでいる.しかしながら,以下の2点について想定通りに進んでおらず,やや遅れていると判断した.1つ目は脳波などの生体情報及び編集履歴からの理解度推定アルゴリズムの構築である.当初本研究の独創的な点として,脳波などの生体情報,及び,編集履歴を使い,学生の理解度を推定することを挙げていた.本年度も昨年度に引き続き,脳波履歴・編集履歴を収集・解析し,学生の理解度推定に有効な特徴量を調査したが,いまだ有効な特徴量は見つかっていない.これらは予備実験を行ったことで,想定以上に難しい問題であることが判明したが,引き続き実験データを収集・解析し,粘り強く調査を行う予定である.2つ目は検証実験が想定より遅れていることである.オンラインジャッジの構築が完了したことで,現在実際にプログラミング演習問題を作成・掲載し,簡単な予備実験を行っている.しかしながら,現在被験者である学生を集めるのが困難であり,検証実験に若干の遅れが生じている.本システムの対象者はプログラミングコンテストで上位獲得を目指し技術向上を図る学生であり,引き続き本システムの重要性を説き,協力者を募る予定である.今後の主な研究の推進方策として次の3つが挙げられる.1つ目はオンラインジャッジサーバのプログラミング演習問題と新たな機能の追加による実用性の向上である.実用的なシステムの実現には大量,かつ,学生のプログラミングスキル向上に有効なプログラミング演習問題が必要不可欠である.まず,多くのプログラミング演習問題を作成し,掲載する.また,実際の運用を行うには,ユーザインタフェースや安全性に関する新たな機能の追加が必要であると考える.また,本研究の独創的な点として挙げている推薦システムの追加も行う.次年度はまずこれら演習問題や新たな機能の追加を行い,より実用的なシステムの完成を目指す.2つ目は生体情報と編集履歴による学生の理解度推定アルゴリズムの構築である.これまでの予備実験からは成果が上がっていないが,引き続きデータの収集と解析を行い,調査を行う.次年度は特に視線情報のデータ収集を行い,学生の理解度推定に有効な特徴量を調査する予定である.3つ目は構築したサーバを学生に利用してもらい,プログラミングスキル向上にどれほど有効か,検証実験をすることである.現在構築したシステムでも実験は可能であり有効な実験データは収集可能である.次年度は協力者を募り,評価実験を進めていく.次年度は以上の3つを並行して進める予定である

  • 有限長解析情報理論と最適化理論による実用高信頼高効率通信に向けた相乗的基礎研究

    Project Year :

    2017.04
    -
    2021.03
     

     View Summary

    本研究では,最終的に,(a)符号化レートや誤り確率の理論限界を,実用的なデータ長や実用的な誤り確率を許して数値として導出する,(b)符号化復号システム全体を大きな最適化問題として定式化し,より実用的な制約のもと(準)最適な符号と復号の組として求める,というように,理論とアルゴリズムの研究を接近させ,融合・発展させること目標としている.本年度は,昨年度に引き続き,(a)に関連した研究として以下の(a')を,(b)に関連した研究として以下の(b')の研究を行った.その結果,(a')に関しては以下の(R-1),(b')に関しては以下の(R-2)の結果を得た.(a') 雑音のある通信路を通しての情報通信の研究である通信路符号化問題において,有限のデータ長に対する符号化レート,誤り確率等の理論限界の導出.(b') 様々な最適化問題とその解法アルゴリズムについての従来研究の整理.また,現状の符号化復号システムにおいてシステムの一部を最適化問題とみなした際の目的関数やヒューリスティクスとして用いられうる量の整理,拡張.(R-1) 構成した符号の有限長での誤り確率の性能を評価する量として,BP閾値と呼ばれる量があるが,対数型空間結合符号およびSpatially Coupled Uneven LDPC Code と呼ばれる新たに提案したうえで,この2つの符号クラスに対して,それぞれ密度発展法という手法による評価を行った.(R-2) 符号構成の目的関数になりうる量として,重み分布と呼ばれる量があるが,``Mt. Fuji’’ Spatially Coupled Code と呼ばれる新たな符号クラスに対して,重み分布を解明し,Covariance Evolution による評価を行った.これにより,符号化復号システム全体の最適化問題としての定式化に近づいた.本研究の最終的な目的は,(a)符号化レートや誤り確率の理論限界を,実用的なデータ長や実用的な誤り確率を許して数値として導出する,(b)符号化復号システム全体を大きな最適化問題として定式化し,より実用的な制約のもと(準)最適な符号と復号の組として求める,ということである.この目的に対して,以下の結果が得られたため,おおむね順調に進展していると判断した.(1) 歪みを許した情報源符号化問題において有限長のデータに対する符号化レートの理論限界を導出することができた.(2) 符号化復号システムにおいて一部を最適化問題とみなした際の目的関数やヒューリスティクスとして用いられうる量の整理,拡張を行うことができた.本年度で得られた成果をもとに,最終的な研究目標の達成に向けて,研究をさらに発展させる.具体的には,研究課題(a)に対しては以下の(i),研究課題(b)に対しては以下の(ii)のアプローチにより研究を実施する予定である.(i)本年度の成果をもとに,より実用に近い仮定のもと,より精密に数値的に誤り確率等が導出できる解析手法を構築する.(ii)本年度に整理した符号化復号システムの一部を最適化問題とみなした問題に対する知識を総合的に考慮して,より実用に近い仮定のもと,符号化復号システム全体を最適化問題として俯瞰し,最適化理論を用いた復号アルゴリズムや符号の探索による符号構成の同時最適化の研究を進めていく.また,アプローチ(i)とアプローチ(ii)を近づけていき,最終的には両アプローチを融合させ相乗的な研究を目指す

  • ベイズ理論による複数目的に対する効率的同時実験を可能にする新たな実験計画法の創成

    Project Year :

    2017.04
    -
    2020.03
     

     View Summary

    本研究の目的は、複数目的に対する効率的同時実験を実施可能な新たな実験計画法を創成し、ベイズ理論による実験計画法におけるデータ収集コスト最小化アルゴリズムを導出することである。今年度は、以下の研究成果を得ることができた。1) フーリエ解析の視点を取り入れることにより、関連する正規直交基底による線形モデルの同時実験を提案し、同時実験によりデータ収集のコストを削減可能であることを示した。2) 前年度までの研究で、実験計画法における従来モデルと本研究で提案する正規直交基底によるモデルの関係が示され、両モデルが行列を用いることにより相互に変換されることが明らかになっている。この変換は、ベイズ理論による枠組みにおいて、解析的な結果の導出に利用することが可能である。今年度は、直交計画を利用する場合、行列積の計算を事前に一度しておくことで、事後分散が容易に得られることを示した。3) 実用面でも広く利用されている直交計画について、複素数を用いる基底関数に適用する場合の性質を明らかにした。さらにこれまでに知られている直交計画の最適性が複素空間でも同様に成り立つことを示した。今年度は、ベイズ理論による複数目的に対しての同時実験を実施可能な実験計画法の枠組みを得ることが主な目的であった。その意味で、おおむね順調に推移していると言える。研究実績の概要で述べた通り、今年度は関連する正規直交基底による線形モデルの同時実験を提案し、また同時実験によりデータ収集のコストを削減可能であることを示すことができた。また今年度は、線形符号と関連が深い直交計画を利用する場合について、行列積の計算を事前に一度しておくことで事後分散が容易に得られることを示した。さらにこれまでに知られている直交計画の最適性が複素空間でも同様に成り立つことを示した。今後は、前年度までに得られた複数目的に対する効率的同時実験を実施可能な枠組みのもとで、ベイズ理論による実験計画法におけるデータ収集コスト最小化アルゴリズムの導出及びその性能評価を行う。ここで実験計画法において、一般には事後確率の導出に時間がかかる場合が多いが、本研究では従来研究とは異なり、モデルを正規直交基底モデルで表現するため、ベイズ理論による機械学習分野で得られている研究成果の多くを利用することが可能である。実験計画法においても機械学習と同様にベイズ線形回帰モデルを定式化し、事後分布の解析的な導出が可能であることが前年度までの本研究成果により明らかとなっている。また、実用面でも広く利用されている直交計画を利用する場合について、事後分散などの性質も明らかとなっている。今後、これらの研究成果をもとに、符号理論及びフーリエ解析の視点を更に取り入れることで、ベイズ理論による実験計画法における直交計画を用いたデータ収集コスト最小化アルゴリズムの導出及び性能評価を行う。最後に、得られた結果について多方面からの考察を行い、他分野への拡張性を明らかにする

  • Mathematical model and evaluation of efficiency and security for next-generation distributed storage systems with high-performance

    Project Year :

    2016.04
    -
    2019.03
     

     View Summary

    We have modeled a high-performance distributed storage system by extending and generalizing the mathematical model of the conventional distributed storage system. In addition, By defining the evaluation criteria of efficiency and security for our model, we have derived a trade-off between efficiency and security. As a result, it is possible to evaluate the performance of high performance distributed storage system considering both efficiency and security. Furthermore, we have presented a construction that optimizes efficiency while maintaining a certain security level

  • Construction of basic theory for next generation recommender system based on spatial analysis and temporal control

    Project Year :

    2016.04
    -
    2019.03
     

     View Summary

    We apply Markov decision processes(MDP) to recommender system with transitions of user classes. A new recommender method which maximizes total reward is proposed under the condition that the true parameters of MDP are known. We also proposed a semi-supervised learning method for recommender system under the condition that the true parameters of MDP are unknown.We apply MDP to questionnaire for a new customer in recommender system. A new questionnaire method which maximizes the total reward is proposed under the condition that the true parameters of MDP are known. We also proposed a semi-supervised learning method for the new customer problem of recommender system under the condition that the true parameters of MDP are unknown

  • A Unified Analysis and Optimization of Information Security System with Probabilistic Components from Viewpoints of Convenience and Safety

    Project Year :

    2013.04
    -
    2016.03
     

     View Summary

    Information security problem with probabilistic components has been formulated by probabilistic models. Theoretical criteria for evaluation such as convenience and safety have been defined clearly and optimal attack or an authentication method has been derived theoretically. Theoretical safety bounds have been evaluated with respect to mathematical models with unified framework for each cipher or security system. A theoretical safety bound or optimality has been clarified with respect to a tradeoff between convenience and safety. New theoretical criteria have been derived for information security systems. Approximation algorithms with high performance for optimal attack or an authentication method have been constructed applying results of studies on problems in related fields such as learning or optimization theory that is formulated by probabilistic models equivalent to our study. Convenience or safety of information security systems has been simulated by applying these algorithms

  • A Study on Unified Attack against Stream Ciphers based on Probabilistic Inference Algorithms

    Project Year :

    2010
    -
    2012
     

     View Summary

    Attack against Stream Cipher reduces to attack against a pseudo random generator. In this study, attack against a pseudo-random number generator was classified based on the choice of target variables and its order and the choice of the local relationships used for attack. We proposed efficient and accurate attack algorithms based on probabilistic inference by studying the problem in unified framework. We revealed the class of pseudo-random number generators that we can attack against based on the acquired knowledge of probabilistic inference so far.

  • Construction of Theory of Digital Analysis

    Project Year :

    2008
    -
    2010
     

     View Summary

    Our research group is composed of scholars working in the areas of discrete mathematics, nonlinear differential equations, information theory and numerical computation. We have organized "Seminar on Digital Analysis" so that members can hold common understanding and insight on the fundamental theories and ideas of digital mathematics. As speakers of this seminar, we have invited 16 researchers who are highly active in the areas of discrete mathematics, mathematical modeling, information theory and numerical computation. We have succeeded in getting common understanding on digital analysis through exciting discussions in each lecture of the seminar.

  • 環境共生型高精度エネルギー予測/高効率エネルギー利用技術

    文部科学省 

    Project Year :

    1998
    -
    2002
     

  • 経営工学における知識情報処理に関する研究

     View Summary

    近年,曖昧性や矛盾を含んだ知識の推論や学習等の,より人間的で複雑な判断をめざした新しい知識情報処理の基礎研究も様々な分野で盛んに行われている.しかしながら従来の論理学を中心としたエキスパートシステムに比べて基礎理論が不完全であり評価基準も不明確であるため得られる結果が理論的に明快に保証されずシステム化が進んでいないのが現状である.本研究ではこれら新しい知識情報処理のために今までの個別的な研究ではなく,人間や組織における判断で取り扱われる情報に着目した統一的な基礎理論の構築をめざしている.そして,その基礎理論上で個々の新しい知識情報処理の分野でネックとなっている基本問題について情報理論,統計理論的尺度から最適なアルゴリズムをいくつか提案し,さらにこのアルゴリズムを応用した知識情報システムの評価を行うことを目的とした.各研究者は主に以下の分担で研究を行った.研究代表者は主として情報の基本的性質に立脚し,情報理論及び符号理論を基本として情報の高信頼化,圧縮の基礎的研究を行うと共に分担研究者それぞれの情報にたいする視点をとりまとめ総合的に知識情報処理問題の考察を行った.研究分担者(石渡)は主として人間や組織における判断に必要となる情報やその判断の機構について基本的視点から考察すると共に,現在の意思決定システムの問題点及び将来像から打開しなければならない本質的問題を抽出し情報的視点から整理を行なった.研究分担者(松嶋)は主として従来型の知識情報処理の基礎理論である論理学だけでは取り扱えない新しい知識情報処理の問題点を整理し,情報理論,統計確率理論,決定理論など情報の基礎理論からの意味づけとモデル化を行った.さらに研究代表者と研究分担者がそれぞれ個別に研究を進めてきた情報に対する視点をまとめ,相互に意見交換することにより,現在及び将来の知識情報処理の基本問題を情報的視点より整理し,新しい基礎理論体系の枠組みの設定を行なった.また,この新しい基礎理論体系の枠組みをさらに洗練化させ,情報の視点から整理された新しい知識情報処理の為に重要ないくつかの基本的問題に対してこの基礎理論のモデルを適用する事により,いくつかの情報的視点から評価基準を設定し,そのもとで最適なアルゴリズムを提示した.また,経営工学やその周辺領域で今後必要となる将来型の知識情報システムにおける上記アルゴリズムの応用についての考察も行った

  • Application of coding techniques to multimedia

     View Summary

    (1) (1)We studied data compression for text in source coding. We theoretically derived difference in code lengths between the codes based on the MDL principle and the Bays codes. Then these results were applied to the model selection problem.(2)We developed an approximately optimal algorithm which reduces complexity in calculation for the Bays code. We applied this algorithm to benchmark test, files, and so on, which gave good results. As data compression for speech, image, and so on, we also proposed an algorithm which applied the trellis codes to the source coding problem, and performed an evaluation experiment for this algorithm.(2) (1)In channel coding, we developed an algorithm on decoding beyond the BCH bound for BCH codes, and also an algorithm on soft decision decoding for block codes. From the viewpoint of complexity in calculation, we found a practical decoding method.(2)For convolutional codes, the relation between the Viterbi decoding method and the syndrome decoding method was elucidated. And these results were applied to the ARQ method. We found a decoding method having excellent performance at low rate.(3)We performed signature analysis by parallel processing as an application of the error-correcting codes to test methods of LSI

  • The Analysis and Design of the Reliable Iterative Decoding Algorithm based on Uncertain Reasoning

     View Summary

    1. We analyzed and generalized our original reasoning algorithm, which is already proposed, based on both Bayesian Network and Graphical Model. Then, we designed a new algorithm for calculating posterior probability on probability models including classes of efficient codes. We also inspected the algorithm in terms of both theoretical and simulating points and then got good results.2. Moreover, we inspected both convergence and accuracy of the algorithm in terms of the differential geometry, machine learning and statistics and then we gave theoretical understandings onto it.3. We defined the generalized posterior probability distribution and made mathematical expression of a problem of the uncertain reasoning given the observed probability distribution. It turned out that the result of our reasoning method has a lot of mathematical meanings that is very important. We improved our algorithm by reducing both the calculation and memory occupation, and by paralyzing its procedures.4. We applied our algorithm on the Extended Junction Graph, which is also our generalized original graph, then proposed the efficient parallel belief propagation algorithm. As a typical application, we applied our algorithm, which is already mentioned in 2., for decoding on Extended Junction Graph constructed by LDPC codes. We examined its natures in terms of both theoretical and experimental aspects.5. We used our Extended Junction Graph for both convolution codes and tail biting codes, then we applied our algorithm mentioned in 2. for decoding both codes. We examined its results from the both mathematical and computational points of view.6. We applied our generalized algorithm for calculating posterior probability, which is already proposed year 2001, to a special model class that has guarantee of convergence. The whole procedure guarantees mathematical meanings as well as both efficient calculation and memory occupation.7. We developed out results in year 2001 on decoding algorithm, then inspected from point of theoretical and experimental view

  • Application and Analysis of Parallel Iterative Algorithms for Calculating Posterior Probability to Practical Codes and Channels

     View Summary

    For its good error correcting ability, Turbo code and Turbo decoding algorithm, which was proposed in the research area of the error correcting codes, have gotten a lot of attention recently as important technology of 21st century's reliable communication and accumulation technique. It have been known that the iterative decoding algorithms which includes Turbo decoding algorithm can be regarded as applications of Belief Propagation algorithm, which was proposed in the research area of the knowledge information processing as uncertain reasoning algorithm. It can be interpreted that the turbo decoding algorithm calculates the approximation of posterior probability effectively.It was obtained that the Extended Junction Graph (EJG) and the parallel iterative algorithm for calculating posterior probability, which are results achieved in "The Analysis and Design of the Reliable Iterative Decoding Algorithm based on Uncertain Reasoning" which is supported by Grant-in-Aid for Scientific Research in 2000-2002, can deal with general probabilistic models compared to existing belief propagation algorithm and retain the higher performance even when the graph has loops (almost of the practical codes have loops).In this research, we have done following,(1)We applied the EJG and the parallel iterative algorithm to the LDPC codes which have some small loops, we sought further efficiency and analyzed the performance.(2)We applied the algorithm to the channel models which have feedback channel such as ARQ Scheme and analyzed the performance.(3)Moreover, we applied the algorithm to the decryption of stream cipher and analyzed the performance

  • Modeling and optimization of the sensor networks based on the multiterminal information theory and decision theory

▼display all

Presentations

  • Optimal Estimating of the Magnitude of the change for Sources with Piecewise Constant Parameters under Bayesian Criterion

    Kairi Suzuki, Akira Kamatsuka, Toshiyasu Matsushima

    Bayes on the Beach 

    Presentation date: 2019.11

  • Bayes Optimal Prediction and Its Approximative Algorithm on Model Including Cluster Explanatory Variables and Regression Explanatory Variables

    Haruka Murayama, Shota Saito, Yuta Nakahara, Toshiyasu Matsushima

    Presentation date: 2019.07

  • Bayes Optimal Classification on Decision Tree Model and Its Approximative Algorithm Using Ensemble Learning

    Nao Dobashi, Shota Saito, Toshiyasu Matsushima

    Presentation date: 2019.07

  • A Note on Mixed Level Experimental Designs Using Augmented Orthogonal Arrays

    Junki Yamaguchi, Koki Kazama, Akira Kamatsuka, Shota Saito, Toshiyasu Matsushima

    Presentation date: 2019.07

  • 攻撃者の目的と背景知識を明確にしたプライバシ保護を考慮した情報公開

    鎌塚明, 宮下有咲, 吉田隆弘, 松嶋敏泰

    みずほ銀行・早稲田大学 学術交流協定締結1周年記念シンポジウム ポスターセッション 

    Presentation date: 2019.07

  • i.p.i.d.情報源におけるベイズ規準の下で最適な変化回数の推定

    鈴木海理, 鎌塚明, 松嶋敏泰

    電子情報通信学会 技術研究報告 IBISML 

    Presentation date: 2019.06

  • Gradient Coding based on LDGM Codes and Stochastic Gradient Descent

    Presentation date: 2019.03

  • 相異なる3元ハフマン符号が構成される2値無記憶拡大情報源の条件に関する一検討

    宮希望, 吉田隆弘, 地主創

    第41回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2018.12

  • Non-Asymptotic and Asymptotic Fundamental Limits of Guessing Subject to Distortion

    Presentation date: 2018.12

  • $(n,k,d,r,t,x,y)_{q}$ LRC符号の最小距離および次元の限界式に関する一考察

    風間皐希, 鎌塚明, 松嶋敏泰

    第41回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2018.12

  • 拡張直交配列を利用した多水準の実験計画法に関する一考察

    山口純輝, 風間皐希, 鎌塚明, 齋藤翔太, 松嶋敏泰

    第41回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2018.12

  • メロディの生成数理モデルを仮定した自動作曲

    西川史織, 中原悠太, 松嶋敏泰

    第41回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2018.12

  • 富士山型空間結合符号に対するCovariance Evolution

    中原悠太, 松嶋敏泰

    第41回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2018.12

  • 攻撃者の目的と背景知識を明確にしたプライバシ保護を考慮した情報公開

    宮下有咲, 鎌塚 明, 吉田隆弘, 松嶋敏泰

    第41回情報理論とその応用シンポジウム予稿集(SITA2018) 

    Presentation date: 2018.12

  • 一般化ラベルノイズの下での分類に関する漸近評価

    安田豪毅, 須子統太, 小林学, 松嶋敏泰

    電子情報通信学会技術研究報告. IBISML,情報論的学習理論と機械学習 

    Presentation date: 2018.11

  • パラメータ未知の一般化ラベルノイズモデルにおける分類法について

    須子統太, 安田豪毅, 堀井俊佑, 小林学

    電子情報通信学会技術研究報告. IBISML,情報論的学習理論と機械学習 

    Presentation date: 2018.11

  • 統計的決定理論に基づいた因果効果の推定法に関する一考察

    堀井俊佑, 須子統太

    電子情報通信学会研究技術報告 

    Presentation date: 2018.11

  • A Note on a Bound on the Rate of a Locally Recoverable Code with Multiple Recovering Sets

    Kazama, Koki, Kamatsuka, Akira, Yoshida, Takahiro, Matsushima, Toshiyasu

    Proceedings of 2018 International Symposium on Information Theory and Its Applications 

    Presentation date: 2018.10

  • Expected Graph Evolution for Spatially “Mt. Fuji” Coupled LDPC Codes

    Nakahara, Yuta, Matsushima, Toshiyasu

    Proceedings of 2018 International Symposium on Information Theory and Its Applications 

    Presentation date: 2018.10

  • On Distance Properties of (r, t, x)-LRC Codes(最新論文紹介セッション)

    風間皐希, 松嶋敏泰

    第7回誤り訂正符号のワークショップ 

    Presentation date: 2018.09

  • Concatenated Spatially Coupled LDPC Codes for Joint Source-Channel Coding(最新論文紹介セッション)

    中原悠太, 松嶋敏泰

    第7回誤り訂正符号のワークショップ 

    Presentation date: 2018.09

  • ランク誤りを考慮した Coded Computation に関する一考察

    風間皐希, 鎌塚明, 松嶋敏泰

    データ科学総合研究教育センター第3回シンポジウム(ポスター発表) 

    Presentation date: 2018.07

  • 潜在変数に階層モデルを仮定したベイズ独立成分分析

    浅葉海, 齋藤翔太, 堀井俊佑, 松嶋敏泰

    電子情報通信学会技術研究報告  電子情報通信学会情報論的学習理論と機械学習研究会(IBISML)

    Presentation date: 2018.03

  • 陽に記述された画像生成モデルに対するベイズ基準のもと最適な可逆符号化

    中原悠太, 松嶋敏泰

    映像情報メディア学会2017年冬季大会講演予稿集  映像情報メディア学会

    Presentation date: 2017.12

  • 微少なアンダーフロー確率を許した可変長intrinsic randomness問題

    吉澤潤, 齋藤翔太, 松嶋敏泰

    第40回情報理論とその応用シンポジウム予稿集  情報理論とその応用学会

    Presentation date: 2017.11

  • 正値の歪み超過確率を許容した可変長有歪み情報源符号化における符号語長のキュムラント母関数

    齋藤翔太, 松嶋敏泰

    第40回情報理論とその応用シンポジウム予稿集  情報理論とその応用学会

    Presentation date: 2017.11

  • ランク誤りを考慮したCoded Computation に関する一考察

    風間皐希, 鎌塚明, 松嶋敏泰

    第40回情報理論とその応用シンポジウム予稿集  情報理論とその応用学会

    Presentation date: 2017.11

  • 最小修復帯域幅を達成する再生成符号に基づく分散ストレージシステムにおける秘匿情報検索に関する一検討

    吉田隆弘, 松嶋敏泰

    第40回情報理論とその応用シンポジウム予稿集  情報理論とその応用学会

    Presentation date: 2017.11

  • 富士山型空間結合符号の重み分布に関する一考察

    中原悠太, 松嶋敏泰

    第40回情報理論とその応用シンポジウム予稿集  情報理論とその応用学会

    Presentation date: 2017.11

  • 統計的決定理論に基づく2次元ウェーブレットパケットを用いた画像修復

    小板橋和也, 中原悠太

    第2回WIRPワークショップ 

    Presentation date: 2017.04

  • 情報理論に基づく無歪み圧縮のための画像の数理モデル化に関する一考察

    中原悠太

    第2回WIRPワークショップ 

    Presentation date: 2017.04

  • A Note on New Schedule of BP Decoding Based on Local Cycle Distributions

    Nakahara, Yuta, Matsushima, Toshiyasu

    Presentation date: 2016.12

  • Variable-Length Lossy Compression Allowing Positive Overflow and Excess Distortion Probabilities

    Presentation date: 2016.12

  • 潜在変数を仮定した非線形回帰モデルおけるベイズ基準のもと最適な予測

    潮田 幹生, 鎌塚 明, 松嶋 敏泰

    第39回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 2016.11

  • 一般化された再生成符号に対する効率的な複数割当法による構成法

    鎌塚明, 吉田隆弘, 松嶋敏泰

    第39回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 2016.11

  • A Note on List Decoding of Codes over Symbol-Pair Read Channel

    Presentation date: 2016.11

  • 富士山型空間結合符号

    中原悠太

    第5回誤り訂正符号のワークショップ 

    Presentation date: 2016.09

  • On the Applications of Weibull Distribution with Nonstationary Scale Parameter

    Presentation date: 2016.03

  • A Note on a Privacy-Preserving Method for Distributed Regularized Logistic Regression

    Presentation date: 2016.01

  • A Note on Unequal Error Protection in Random Network Coding

    Presentation date: 2016.01

  • A Note on the Optimal Data Prediction under Bayes Criterion for Multidimensional Linear Regression Models Assuming Latent Variables

    Presentation date: 2016.01

  • A Study on Soft Decision Decoding for Array-Error Channel

    Presentation date: 2016.01

  • A Study on Message Passing Algorithm for Counting Short Cycles in Sparse Bipartite Graphs

    Presentation date: 2016.01

  • A Note on the Computational Complexity Reduction Method of the Optimal Prediction under Bayes Criterion in Semi-Supervised Learning

    Presentation date: 2016.01

  • A Note on the Linear Programming Decoding of Linear Codes for Symbol Pair Read Channel

    Presentation date: 2015.11

  • A Study of Regenerating Codes with Generalized Conditions of Reconstruction and Regeneration

    Presentation date: 2015.11

  • A Note on the Decoding of Polar codes using Search Algorithm

    Presentation date: 2015.11

  • 一般情報源に対するSmooth最大エントロピーを用いた可変長符号化の達成可能オーバーフローしきい値について

    第38回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 2015.11

  • On the Credible Interval Prediction of Web Traffic based on the Bayes Decision Theory

    Presentation date: 2015.11

  • On decoding of Polar codes using search algorithm

    Presentation date: 2015.09

  • Linear Programming Decoding of Binary Linear Codes over Symbol-Pair Read Channels

    Presentation date: 2015.09

  • Variations of the Strong Converse Theorem on the Intrinsic Randomness Problem for General Sources

    Presentation date: 2015.07

  • On the Bayes Optimal Prediction of Time Series under a Nonstationary Weibull Distribution

    Presentation date: 2015.07

  • Another Representation on the Second-Order Achievable Rate Region of Slepian-Wolf Coding Problem for General Sources

    Presentation date: 2015.03

  • A Note on the Linear Bandit Problem with Multiple Parameters

    Presentation date: 2014.12

  • プライバシー保護機能を持つ分散型正則化最小二乗法について

    第37回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 2014.12

  • Universality of Spatially Coupled Punctured LDPC Codes for Decode-and-Forward in Erasure Relay Channels

    Presentation date: 2014.12

  • Evaluation of the Minimum Overflow Threshold of Bayes Codes for Non-Stationary Sources

    Presentation date: 2014.12

  • A Note on the Effectiveness of Unlabeled Data for Prediction Error in Semi-Supervised Learning

    Presentation date: 2014.12

  • The Optimal Prediction under Bayes criterion for Linear Regression Models with Latent Variables

    Presentation date: 2014.12

  • Asymptotic Normality and LIL of the Codeword Length of Bayes Codes for Stationary Ergodic Markov Sources

    Presentation date: 2014.12

  • A Note on Attack Against Nonlinear Combiner Generator Using Sum - Product Algorithm

    Presentation date: 2014.11

  • A Note on Improvement in the Rate of a Prediction Error of AdaBoost in Pattern Recognition

    Presentation date: 2014.09

  • How to Observe Probability Distribution of Self-Information as viewed under different resolutions of $n^{-1}$ and $n^{-1/2}$

    Presentation date: 2014.07

  • A study of a MDP - based recommender system when the class of user who is recommended items is unknown

    Presentation date: 2014.07

  • A Note on Optimal Control System for Selective - Repeat Hybrid SR - ARQ with a Finite Length Buffer

    Presentation date: 2014.07

  • A Note on the Correlated Multiple Matrix Completion based on the Convex Optimization Method

    Presentation date: 2013.11

  • Asymptotics of Consistent Estimator - based Prediction for Semi-supervised Learning

    Presentation date: 2013.11

  • An Approximation of Bayes Prediction for Linear Regression Models

    Presentation date: 2013.11

  • Relationship Between the Overflow Probability of Variable-length Coding and the Error Probability of Fixed-length Coding

    Presentation date: 2013.11

  • Asymptotics of Bayesian Estimation for Hierarchical and Misspecied Models

    Presentation date: 2013.10

  • Evaluation of Minimum Overflow Threshold for Bayes Codes

    Presentation date: 2013.10

  • A Note on Construction of Parallel Concatenation Using Polar Codes

    Presentation date: 2013.10

  • A Consideration on the Storage and the Repair-bandwidth for Generalized Class of Regenerating Codes based on Leakage of Share

    Presentation date: 2013.10

  • Bayesian Prediction for Multivariate Polynomial Regression with Unknown Order

    Presentation date: 2013.10

  • Lossless Image Data Compression Based on Regression Trees

    Presentation date: 2013.10

  • An Approximation of Bayes Prediction for Linear Regression Models

    Presentation date: 2013.10

  • The Empirical Bayes Forecasting of Web Traffic on the Non-Stationary Poisson Distribution

    Presentation date: 2013.10

  • Iterative Multiuser Joint Decoding based on Augmented Lagrangian Method

    Presentation date: 2013.09

  • A Consideration on Modeling and Optimality of Regenerating Codes Considering Security of Shares

    Presentation date: 2013.07

  • プライバシー保護を目的とした線形回帰モデルにおける事後確率最大推定量の分散計算法について

    NAKAI Akito, SUKO Tota, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 2013.03

  • A Consideration on Minimum Storage Regenerating Codes for Functions

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. WBS, ワイドバンドシステム 

    Presentation date: 2013.03

  • プライバシー保護を目的とした線形回帰モデルにおける事後確率最大推定量の分散計算法について

    NAKAI Akito, SUKO Tota, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 2013.03

  • A Consideration on Minimum Storage Regenerating Codes for Functions

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. WBS, ワイドバンドシステム 

    Presentation date: 2013.03

  • An Extended Privacy Preserving Regression Analysis

    Presentation date: 2012.12

  • A Note on Tree Model-Based Compressed Sensing Based on Augmented Lagrangian Method

    Presentation date: 2012.12

  • Linear Time ADMM Decoding for LDPC Codes over MIMO Channels

    Presentation date: 2012.12

  • A Privacy Preserving Distributed Calculation Method of Least-squares Estimator for Linear Regression Models

    Presentation date: 2012.11

  • An optimal prediction method using hierarchical N-gram based on Bayes decision theory

    Presentation date: 2012.09

  • Linear Programming Decoding for Multiple Access Channel based on Decomposition Methods

    Presentation date: 2012.09

  • Variable Order Transition Probability Markov Decision Process for the Recommendation System

    Presentation date: 2012.05

  • ベイズ決定理論にもとづく階層N グラムを用いた最適予測法と日本語入力支援技術への応用

    言語処理学会第18回年次大会(NLP2012)発表論文集 

    Presentation date: 2012.03

  • マルコフモデルによる自動分類に対する分類誤り確率の推定

    情報処理学会研究報告 

    Presentation date: 2012.03

  • Information Theory and Learning Theory

    Presentation date: 2012.03

  • Universal Markov Decision Process Designed by Bayes Decision Theory for the Recommendation System

    Presentation date: 2012.03

  • 実験計画法に適した直交配列の線形計画限界

    情報処理学会第74回全国大会 

    Presentation date: 2012.03

  • A Note on Analysing LDPC Codes for Correcting a Burst Erasure

    Presentation date: 2012.01

  • An Error Probability Analysis of the Text Classification Using the CTW Algorithm

    KOBAYASHI Manabu, GOTO Masayuki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. NLP, 非線形問題 

    Presentation date: 2011.11

  • An Error Probability Analysis of the Text Classification Using the CTW Algorithm

    KOBAYASHI Manabu, GOTO Masayuki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. NLP, 非線形問題 

    Presentation date: 2011.11

  • 次数が未知の状態遷移確率を仮定したマルコフ決定過

    第34回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 2011.11

  • A Study on Key Estimation of Stream Cipher based on Probabilistic Inference Algorithm

    Presentation date: 2011.07

  • A study of statistical modeling of authentication using PUF

    Presentation date: 2011.07

  • Asymptotics of Bayesian prediction for misspecified models

    Presentation date: 2011.07

  • A consideration on unconditionally secure key distribution schemes considering leakage of secret information of key distribution centers

    Presentation date: 2011.03

  • Lower bounds on the memory size of entities for a ramp key distribution scheme and an optimal construction

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

    Presentation date: 2011.01

  • Disk Allocation Methods for Cartesian Product Files by using Unequal Error Protection Codes

    SAITO Tomohiko, INAZUMI Hiroshige, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. IT, 情報理論 

    Presentation date: 2011.01

  • Lower bounds on the memory size of entities for a ramp key distribution scheme and an optimal construction

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

    Presentation date: 2011.01

  • Disk Allocation Methods for Cartesian Product Files by using Unequal Error Protection Codes

    SAITO Tomohiko, INAZUMI Hiroshige, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. IT, 情報理論 

    Presentation date: 2011.01

  • A Note on the Inference Algorithm on the Factor Graph based on the Linear Programming

    HORII Shunsuke, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. IT, 情報理論 

    Presentation date: 2011.01

  • A Note on the Inference Algorithm on the Factor Graph based on the Linear Programming

    HORII Shunsuke, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. IT, 情報理論 

    Presentation date: 2011.01

  • Maximum likelihood detection for DS-CDMA using Gr\{o}bner bases

    Presentation date: 2010.11

  • 複数の相関のある情報源に対するベイズ符号化について

    第33回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 2010.11

  • ,Achievable Condition in Resolvability Problem for Mixed Sources

    Presentation date: 2010.11

  • Linear Programming Decoding of Binary Linear Codes for Multiple-Access Channel

    HORII Shunsuke, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. IT, 情報理論  (!) 

    Presentation date: 2010.09

  • Linear Programming Decoding of Binary Linear Codes for Multiple-Access Channel

    (!) 

    Presentation date: 2010.09

  • Linear Programming Decoding of Binary Linear Codes for Multiple-Access Channel

    (!) 

    Presentation date: 2010.09

  • A Study on Key Estimation Attacks to Stream Cipher based on Statistical Decision Theory

    Presentation date: 2010.03

  • Universal Source Coding under an Unknown Alphabet

    Presentation date: 2009.12

  • A Note on Noiseless Separate Source Coding Theorems for Correlated Sources

    Presentation date: 2009.09

  • A Note on the Fixed-Length Source Coding Theorems for Sources with Side Information

    Presentation date: 2009.09

  • An Efficient Algorithm for the Bayes Codes using an Unlimited Depth Context Tree

    2008 SIP (IASTED Signal and Image Processing) 

    Presentation date: 2008.08

  • A Note on the Weak Universal Joint Source - Channel Coding

    Presentation date: 2008.01

  • A Note on Multiuser Detection Algorithms for CDMA based on the Belief Propagation Algorithm

    Presentation date: 2008.01

  • A Note on Multiuser Detection Algorithms for CDMA based on the Belief Propagation Algorithm

    Presentation date: 2008.01

  • 開始と締切りが設定されたシステムの到着モデルとその応用

    オペレーションリサーチ学会 待ち行列研究部会 (招待講演) 

    Presentation date: 2007.11

  • On the Condition of epsilon-Achievable Overflow Thresholds for the Parametric Compound Sources

    Presentation date: 2007.05

  • On the Condition of epsilon-Achievable Overflow Thresholds for the Parametric Compound Sources

    Presentation date: 2007.05

  • Shortened and Concatenated Collusion-Secure Codes for Digital Fingerprinting

    Presentation date: 2007.05

  • A performance of the LZ78 code for Non-prefix-free Word-valued source

    Presentation date: 2007.01

  • Performance Analysis of EM Decoding Algorithm for HMM Channels

    Presentation date: 2006.11

  • A Note on the overflow probability of lossless codes

    Presentation date: 2006.11

  • Iterative decoding and detection techniques (invited talk)

    IEEE ITW06 

    Presentation date: 2006.10

  • On Error Exponents for Variable Size List decoder using the Viterbi Algorithm with Likelihood Ratio Testing

    Presentation date: 2005.11

  • Collusion-Secure Codes for Fingerprinting based on Finite Geometries

    Presentation date: 2005.11

  • A Study of Reliability Based Hybrid ARQ Schemes Using a Recursive Systematic Convolutional Code

    Presentation date: 2005.09

  • A Relation between Erasure Correcting and Error Correcting of Low-Density Parity-Check Codes

    Presentation date: 2005.09

  • インターネットトラヒックのポアソン分布の密度パラメータが時間変動する時系列モデルを用いた解析に関する一考察

    第4回情報科学技術フォーラム(FIT2005) 

    Presentation date: 2005.09

  • Matrix Clustering Method Achieving Specific Accuracy by Modified Apriori Algorithm

    Presentation date: 2005.06

  • A Note on Construction of Nonlinear Unequal Orthogonal Arrays from Error-Correcting Codes

    Presentation date: 2005.05

  • A Classification of the Probabilistic Reasoning given Distribution Evidence and Kullback-Leibler Information

    Presentation date: 2005.04

  • Universal Coding Algorithm for Side Information Context Tree Models

    Presentation date: 2004.12

  • Word segmentation of the sequences emitted from a word-valued source

    Presentation date: 2004.12

  • Universal Coding Algorithm for Side Information Context Tree Models

    第27回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2004.12

  • Bayes Universal Coding Algorithm for Side Information Context Tree Models

    Technical Program of the 4th Asia 

    Presentation date: 2004.08

  • Tailbiting 畳込み符号の復号アルゴリズムに関する一考察

    電子情報通信学会技術研究報告IT2004 

    Presentation date: 2004.07

  • ユニバーサル通信路符号化法における通信路容量について

    電子情報通信学会技術研究報告 

    Presentation date: 2004.03

  • A Note on Models of a Non-Prefix-Free Word-Valued Source

    Presentation date: 2003.12

  • 区間で一定なパラメータを持つ情報源におけるベイズ符号化法について

    第26回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2003.12

  • 衝突困難ハッシュ関数の安全性について

    第26回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2003.12

  • ベイズ決定理論に基づく予測における近似手法について

    第26回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2003.12

  • Parallel Propagation Algorithms for Tailbiting Convolutional Codes

    第26回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2003.12

  • 決定木モデルにおける予測アルゴリズムについて

    電子情報通信学会技術研究報告COMP2003 

    Presentation date: 2003.08

  • 相関のある時系列の状態空間によるモデル化と予測

    電子情報通信学会技術研究報告IT2003 

    Presentation date: 2003.07

  • 畳み込み符号の並列復号アルゴリズムの性能評価に関する一考察

    電子情報通信学会技術研究報告IT2003 

    Presentation date: 2003.07

  • ポアソン分布に従う非定常な時系列のモデル化に関する一考察

    電子情報通信学会技術研究報告IT2003 

    Presentation date: 2003.07

  • Generalized Posterior Probability and Its Calculation

    Technical Program of the 3rd Asia 

    Presentation date: 2003.06

  • Properties of a Word-valued Source with a Non-prefix-free Word Set

    Presentation date: 2003.05

  • 語頭条件を満たさない単語集合をもつWord-Valued Sourceの性質について

    電子情報通信学会技術報告IT2003 

    Presentation date: 2003.05

  • A Note on Robust Pattern Recognition based on Bayes Decision Theory

    Presentation date: 2002.12

  • Prediction Algorithm for Extended Hierarchical Models

    Presentation date: 2002.12

  • 単語単位で系列を出力する情報源の性質について

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • 誤り訂正符号を用いた直行計画の構成法に関する一考察

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • 拡張された階層モデルにおける予測アルゴリズムについて

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • リスト復号に対する判定帰還方式LR+THの誤り指数について

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • 一般化事後確率とその計算法

    電子情報通信学会技術研究報告IT2002 

    Presentation date: 2002.12

  • ベイズ決定理論に基づくロバストなパターン認識に関する一考察

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • 多端子情報源符号化に基づいた分散協調問題の定式化

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • Calculation of Generalized Posterior Distribution on Junction Graphs

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • On the Channel Capacity of Universal Channel Coding

    第25回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2002.12

  • ブール関数の逐次実験計画を用いた学習に関する一考察

    電子情報通信学会技術研究報告COMP2002 

    Presentation date: 2002.11

  • Computational Source Information Model and its Algorithm

    102 

    Presentation date: 2002.09

  • 「インターネットを用いたゼミと研究指導」実用化報告

    2002PCカンファレンス論文集 

    Presentation date: 2002.08

  • A study of calculating channel capacity for (2,2;2)-Multiple Access Channel

    Presentation date: 2002.07

  • 計算論的学習と情報圧縮に関する一考察

    電子情報通信学会技術研究報告IT2002 

    Presentation date: 2002.07

  • (2,2,2)-MACの通信路容量を求める手法に関する一考察

    電子情報通信学会技術研究報告IT2002 

    Presentation date: 2002.07

  • 低密度パリティチェック符号の復号アルゴリズムに関する一考察

    電子情報通信学会技術研究報告SST2001 

    Presentation date: 2002.03

  • ブロックターボ符号に対するインタリーバの構成法と最小距離

    第24回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2001.12

  • 多端子情報理論に基づく分散協調問題について

    第24回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2001.12

  • 「インターネットを用いた研究活動支援システム」システム構成と評価

    第13回日本経営工学会秋季大会 

    Presentation date: 2001.11

  • ベイズ決定理論による定式化のもとで直交計画を用いたブール関数の学習に関する一考察

    電子情報通信学会技術報告COMP2001 

    Presentation date: 2001.11

  • 形式言語と圧縮に関する一考察

    電子情報通信学会技術報告IT2001 

    Presentation date: 2001.11

  • 状態空間モデルを用いた時系列解析に関する一考察−モンテカルロフィルタにおけるサンプリング方法について−

    日本経営工学会秋季研究大会予稿集 

    Presentation date: 2001.11

  • ターボ符号,LDPC符号の復号アルゴリズム

    ベイジアンネットチュートリアル講演論文集 

    Presentation date: 2001.07

  • ブロックターボ符号の生成行列と性能評価

    電子情報通信学会技術研究報告IT2001 

    Presentation date: 2001.07

  • 多端子モデルに基づく分散協調問題の定式化について

    電子情報通信学会技術研究報告IT2001 

    Presentation date: 2001.07

  • パラメータが時間変化する情報源とその符号化に関する一考察

    電子情報通信学会技術研究報告IT2001 

    Presentation date: 2001.07

  • 微分幾何から見た確率推論と復号

    確率伝搬に基づく復号法と符号ワークショップ 

    Presentation date: 2001.03

  • ウェーブレット・パケットを用いた雑音除去におけるベイズ法の応用に関する一考察

    電子情報通信学会技術研究報告DSP2000 

    Presentation date: 2000.12

  • フーリエ変換を用いたブール関数の学習に関する一考察

    電子情報通信学会技術研究報告COMP2000 

    Presentation date: 2000.11

  • 不確実性を含む演繹推論に関する一考察

    第23回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2000.10

  • ユニバーサル符号における固定長符号の誤り率と可変長符号のオーバーフロー確率について

    第23回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2000.10

  • 尤度比検定を用いた木符号の復号法について

    第23回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2000.10

  • An Iterative Caluculation Algorithm for Posterior Probability

    第23回情報理論とその応用シンポジウム予稿集 

    Presentation date: 2000.10

  • Methematical Models in Information-Based Induction Science

    Presentation date: 2000.07

  • 確率推定誤差を考慮に入れた算術符号化アルゴリズム

    電子情報通信学会技術研究報告IT2000 

    Presentation date: 2000.07

  • 不確実な知識の演繹推論における二項述語への拡張に関する一考察

    人工知能学会全国大会(第14回)論文集 

    Presentation date: 2000.07

  • Ziv-Lempel符号の冗長度に関する一考察

    電子情報通信学会技術研究報告IT2000 

    Presentation date: 2000.07

  • 符号長の分散とオーバーフロー確率について

    第22回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • 不確実な知識の演繹推論アルゴリズムに関する一考察

    第22回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • 単語単位で系列を出力する情報源

    第22回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • 信頼度情報に基づく置換生成行列を用いた最尤復号法の効率化について

    第22回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • 質問からのブール関数の学習における学習戦略を求めるアルゴリズム

    第23回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • コスト付き情報源符号化定理について

    第22回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • Viterbi アルゴリズムを用いた判定帰還方式における再送要求によるリスタートとスループットについて

    第22回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • A Geometrical Interpretation on the Calculation of Posterior Probability and Turbo Decoding

    第23回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1999.12

  • Iterative Algorithms for Calculating Probability and Decoding

    Presentation date: 1999.09

  • 実用化せまるturbo符号・復号の理論と実際

    電気情報通信学会ソサイエティ大会 

    Presentation date: 1999.09

  • On the Error Probability of Model Selection for Classisfication

    1999 Workshop on Information-Based Induction Science 

    Presentation date: 1999.08

  • Compression and decoding of an inverted file by using syndrome

    Presentation date: 1999.07

  • CELP音声符号化方式のランダム駆動信号に関する一考察

    電子通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1999.07

  • 誤り訂正符号の復号アルゴリズム

    電気情報通信学会総合大会 

    Presentation date: 1999.04

  • Soft-Decision Decoding Methods using Algebraic Decoding Algorithm

    Presentation date: 1999.03

  • 木符号におけるリスト復号法を用いた判定帰還方式について

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • 不確実性を伴う診断型エキスパートシステムのモデル化と推論法について

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • 不確実な知識の推論における欠測データの取り扱いに関する一考察

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • 適合フィードバックによる情報検索に関する一考察

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • 近似的なベイズ学習と学習可能性について

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • 音声認識における Hidden Markov Model のパラメータ推定に関する一考察

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • モデル族の部分集合に基づく予測方法について

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • Extended Stochastic Complexity の漸近式について

    第21回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1998.12

  • 帰納・演繹推論と予測−決定理論による学習モデル

    1998年情報論的学習理論ワークショップ予稿集/情報理論とその応用学会 

    Presentation date: 1998.07

  • The Generalization of Bayesian Network's Deductive Method

    信学技報/電子情報通信学会 

    Presentation date: 1998.07

  • 対数線形モデルを用いた不確実な知識の推論法について

    人工知能学会全国大会(第12回)論文集/人工知能学会 

    Presentation date: 1998.06

  • マルコフ決定過程の計算アルゴリズムについて

    人工知能学会全国大会(第12回)論文集/人工知能学会 

    Presentation date: 1998.06

  • On performance of prediction using side information

    Presentation date: 1998.05

  • 事後確率密度の漸近正規性

    平成10年度春季大会予稿集/日本経営工学会 

    Presentation date: 1998.05

  • 遺伝的アルゴリズムとK-means法を用いた非階層クラスタリング

    平成10年度春季大会予稿集/日本経営工学会 

    Presentation date: 1998.05

  • 階層モデル族のモデル選択における選択誤り率について

    信学技報/電子情報通信学会 

    Presentation date: 1998.01

  • Type for hierarchical probabilistic models

    Presentation date: 1997.12

  • A study of improving the Ziv - Lempel78 Code from the view point of Bayes Code

    Presentation date: 1997.12

  • 木構造型モデル族のモデル選択法に関する一考察

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • 不確実性を含むデータの統合に関する一考察

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • 直交表現された仮説の学習に関する一考察

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • 質問からの学習における予測誤りに関する一考察

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • 混合分布の近似とその性能について

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • 決定を考慮したベクトル量子化法の提案

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • 階層的確率モデルにおけるタイプについて

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • ベイズ符号化法におけるメモリの問題点に関する一考察

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • ベイズ符号の視点からのZiv-Lempel78符号の改良に関する一考察

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • BCH限界を超える復号アルゴリズムを用いたChase復号法の計算量低減

    第20回情報理論とその応用シンポジウム予稿集 

    Presentation date: 1997.12

  • On Decoding Method Beyond the BCH Bound

    Presentation date: 1997.07

  • 木構造モデル族の学習、予測アルゴリズムに関する一考察

    信学技報/電子情報通信学会 

    Presentation date: 1997.07

  • 不均一誤り訂正符号の復号法に関する一考察

    信学技報/電子情報通信学会 

    Presentation date: 1997.07

  • ベイズ統計学に基づく計算論的学習モデルと学習可能性

    信学技報/電子情報通信学会 

    Presentation date: 1997.07

  • ベイズ決定理論に基づく統計的モデル選択について

    信学技報/電子情報通信学会 

    Presentation date: 1997.07

  • トレリス符号を用いた有歪みデータ圧縮の一考察

    信学技報/電子情報通信学会 

    Presentation date: 1997.07

  • 統計的決定理論によるデータ検索の定式化と最適化

    人工知能学会全国大会(第11回)論文集/人工知能学会 

    Presentation date: 1997.06

  • 属性値の類似度を用いた概念学習の効率化

    春季大会予稿集/日本経営工学会 

    Presentation date: 1997.05

  • 属性のクラスタを用いた概念学習の効率化

    経営情報学会 

    Presentation date: 1997.05

  • ハッシュ技法によるデータ探索の数学的モデル化,及び探索効率の漸近的評価

    信学技報/電子情報通信学会 

    Presentation date: 1997.05

  • ClarkeとBarronのBayesian Asymptoticsについて

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1997.01

  • A Study of An Automatic - Repeart - reQuest Strategy for Non-Stationary Channels

    Presentation date: 1996.12

  • On Soft-Decision Decoding using Error Correcting decoder Beyond the BCH Bound

    Presentation date: 1996.12

  • Codelength of Sufficient Statistic Code

    Presentation date: 1996.12

  • 異なる事前分布を持つ場合の MDL 原理に基づく符号とベイズ符号の符号長に関する一考察

    第19回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1996.12

  • On a Weight Structure for On-lineLearning and its Mistake Bound

    Presentation date: 1996.12

  • Machine Learning Ensembles of Hypotheses

    Presentation date: 1996.12

  • A Note on Variable to Fixed-Length Codes for Markov Sources

    Presentation date: 1996.12

  • A Study on Multiple-Input Signatusre Registers in LSI Self - Testing

    Presentation date: 1996.12

  • Bayes Coding for Data Compression

    Presentation date: 1996.12

  • A study of Ziv - Lempel Algorithm from the view point of Bayes Code

    Presentation date: 1996.07

  • On Complexity of Decoding Beyond the BCH Bound Using Berlekamp-Massey Algorithm : An Application for Soft-Decision Decoding Using Algebraic Decoder

    Presentation date: 1996.07

  • On Bayes Coding in the case of limited memories

    Presentation date: 1996.07

  • A Study on Error Probability of Statistical Model Selection

    Presentation date: 1996.07

  • A Note on Learning Uncertain Concept with Membership Queries

    Presentation date: 1996.06

  • An Algorithm Learning Probabilistic Rules

    Presentation date: 1996.06

  • An Analysis on Difference of the Codelengths between Codes Based on MDL Principle and Bayes Codes

    Presentation date: 1996.05

  • On coded AQR schemes cumulating the rejected sequences for very noisy channel

    Presentation date: 1996.05

  • LSI自己検査における並列シグネチャ解析器の構成

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1996.04

  • 構成的帰納論理プログラミングに関する一考察

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1996.02

  • 巡回符号の並列符号器・復号器に関する検討

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1995.11

  • On Soft Decision Decoding for Reed-Solomon Codes

    Presentation date: 1995.10

  • On Reasoning Methods for Uncertain Knowledge from Noisy Examples

    Presentation date: 1995.10

  • On Universal Codes for FSMX Sources

    Presentation date: 1995.10

  • 矛盾を含む知識の取り扱いについての一考察

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • 不確実な知識の表現法と推論法について

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • 多重符号化を用いた判定帰還方式について

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • 選言的論理プログラムの効率的推論法について

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • 巡回符号の並列符号化器に関する検討

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • 質問を許す概念学習に関する一考察

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • 観測雑音を考慮した不確実な知識の推論法について

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • ユニバーサル情報源符号化アルゴリズムに基づく事後確率推定アルゴリズム

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • ベイズ決定理論に基づくデータ解析に関する一考察

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • Reed-Solomon符号の軟判定一復号法

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • On the Optimum Learning Algorithm for Boolean Functions and the Learnability based on Bayesian Statistics

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • FSMX情報源のユニヴァーサル符号について

    第18回情報理論とその応用シンポジウム予稿集/情報理論とその応用学会 

    Presentation date: 1995.10

  • A Note on Learning Theory for Probabilistic Models

    Presentation date: 1995.09

  • Note on Maximum Likelihood Decoding of Linear Block Codes

    Presentation date: 1995.07

  • A Study on Parameter Estimation and Ziv-Lempel Code

    Presentation date: 1995.07

  • 未知パラメータを含むマルコフ決定過程に関する一考察

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1995.07

  • ベイズ符号を応用した疑似乱数の検定法

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1995.07

  • ベイズ符号による情報源圧縮に関する一考察

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1995.07

  • Optimum Learning Rules for Boolean functions based on Bayesian Statistics

    Presentation date: 1995

  • On the erasure option for the Viterbi algorithm and ARQ schemes

    Presentation date: 1994.12

  • A Two-Stage-Procedure for Universal Coding

    Presentation date: 1994.12

  • A Study on an Application of Neural Network to the {0,1} Classification Problem

    Presentation date: 1994.12

  • A Note on Update of Uncertain Knowledge

    Presentation date: 1994.07

  • A Note on the Construction Method of Decision Trees

    Presentation date: 1994.07

  • On the Complexity of Hypothesis Space and the Sample Complexity for Machine Learning

    Presentation date: 1994.07

  • Inductive and Deductive Inference based on Information Theory

    Presentation date: 1994.07

  • An Efficient Procedure of Inductive Learning for PROLOG

    Presentation date: 1994.06

  • An Inference from a Contradictory Knowledge

    Presentation date: 1994.06

  • 確信度と信頼率を用いた不確実な知識の表現と推論

    人工知能学会全国大会(第8回) 

    Presentation date: 1994.06

  • On coded ARQ schemes cumulating the rejected sequences

    Presentation date: 1994.05

  • On an ARQ using VA with cumulating the rejected sequences

    Presentation date: 1993.10

  • On a Machine Learning Model Based upon Decision Theory

    Presentation date: 1993.10

  • On block coded ARQ schemes cumulating the rejected information

    Presentation date: 1993.07

  • On Inductive Inference and Deductive Inference for Uncertain Knowledge from the view point of Information Theory

    Presentation date: 1992.09

  • Minmax Redundancy Distortionless Codes Constructed from A Class of Bayes Codes

    Presentation date: 1991.12

  • On decision criterion of decision feedback schemes

    Presentation date: 1991.12

  • Concept Learning via a finite number of Membership Queries

    Presentation date: 1991.12

  • On Study of Learning Algorithm for Monotone Conjunctions

    Presentation date: 1991.12

  • A Note on Hypothesis Selection in Hypothesis - based Reasoning

    Presentation date: 1991.01

  • An Inductive Inference Procedure to Minimize Prediction Error

    Presentation date: 1991.01

  • On decison feed - back scheme using Viterbi algorithm

    Presentation date: 1991.01

  • 有限系列に関する平均符号長を考慮したユニバーサル情報源符号化について

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1991

  • A Study of Understanding Simile Using Sense Scales

    Presentation date: 1991.01

  • On a formulation of inference from the view point of information theoery and its application

    Presentation date: 1989.07

  • A Study on Realization of Universal Data Compression

    Presentation date: 1988.12

  • A Strategy on Feature Ordering and Stopping Rule with Maximizing Mutual Information

    Presentation date: 1987.11

  • Notes On the Decoding Time for Concatenated Codes

    Presentation date: 1987.11

  • ,On Algorithms with Information Measure for Production System Strategie

    Presentation date: 1987.11

  • A Stady of Discrete Data Analisis by Information Resolution

    Presentation date: 1986.10

  • A Study on Stopping Rules for Pattern Recognition

    Presentation date: 1986.10

  • 離散データの情報構造モデルに関する一考察

    電子情報通信学会技術研究報告/電子情報通信学会 

    Presentation date: 1986

  • A Consideration on Modeling and Optimality of Regenerating Codes Considering Security of Shares

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

  • Hierarchical Multi-label Classification on Statistical Decision Theory

    YAMAMOTO Kiyohito, SUKO Tota, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IBISML, 情報論的学習理論と機械学習 

  • A Privacy Preserving Distributed Calculation Method of Least-squares Estimator for Linear Regression Models

    SUKO Tota, HORII Shunsuke, KOBAYASHI Manabu, GOTO Masayuki, MATSUSHIMA Toshiyasu, HIRASAWA Shigeichi

    電子情報通信学会技術研究報告. IBISML, 情報論的学習理論と機械学習 

  • Linear Programming Decoding for Multiple Access Channel based on Decomposition Methods

    HORII Shunsuke, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

  • Asymptotics of Bayesian prediction for misspecified models

    MIYA Nozomi, SUKO Tota, YASUDA Goki, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

  • A study of statistical modeling of authentication using PUF

    ISHII Satoru, YOSHIDA Takahiro, HORII Shunsuke, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

  • A Study on Key Estimation of Stream Cipher based on Probabilistic Inference Algorithm

    IIKUBO Yuji, HORII Shunsuke, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. IT, 情報理論 

  • A consideration on unconditionally secure key distribution schemes considering leakage of secret information of key distribution centers

    YOSHIDA Takahiro, JINUSHI Hajime, MATSUSHIMA Toshiyasu

    電子情報通信学会技術研究報告. WBS, ワイドバンドシステム 

  • AK-2-4 Information Theory and Learning Theory

    MATSUSHIMA Toshiyasu

    電子情報通信学会総合大会講演論文集 

  • Berlekamo-Masseyアルゴリズムを用いたBCH限界を超える復号法の計算量について(共著)

    電子情報通信学会論文誌 

  • A Study of the Decision of Control Parameters for Adaptive Automatic-Repeat Request Strategy

    Electronics and Communications in Japan 

  • Fast Correlation Attackの改良法に関する一考察

  • ID情報に基づくランプ型分散鍵配送方式につい て

    第27回情報理論とその応用シンポジウム予稿集 

  • Small-Loopを含む低密度パリティ検査(LDPC)符 号の復号に関する研究

    第27回情報理論とその応用シンポジウム予稿集 

  • ランプ型鍵配送方式について

    電子情報通信学会技術報告 

  • 階層モデルにおけるベイズ予測の漸近評価に関する一考察

    第27回情報理論とその応用シンポジウム 

  • 外れ値データの発生を考慮にいれた回帰モデルにおけるベイズ予測法について

    2004年情報論的学習理論ワークショップ(IBIS2004)予稿集 

  • 確率推論とグラフ伝播復号アルゴリズム

    ベイジアンネットワークセミナー2004 

  • 確率推論とその周辺

    人工知能学会全国大会 

  • 区間で一定なパラメータを持つ非定常情報源におけるベイズ符号の冗長度について

    第27回情報理論とその応用シンポジウム予稿集 

  • 区間で定常なパラメータを持つ非定常情報源におけるベイズ符号の冗長度について

    電子情報通信学会技術報告IT2004-22 

  • 誤り訂正符号を利用した直交計画の構成法に関す る一考察 〜逐次実験に適した直交計画について〜

    第27回情報理論とその応用シ ンポジウム予稿集 

  • 質問学習と逐次実験計画の関係に関する一考察

    電子情報通信学会研究技術報告 AI2003-63 

  • 質問学習に直交計画を用いた場合の予測アルゴリズムに関する一考察

    人工知能基本問題研究会資料 SIG-FPAI-A402-12 

  • 電子透かしにおける秘匿容量計算計算アルゴリズム

    電子情報通信学会技術報告IT2005-94 

  • BW変換を用いたユニバーサル符号化アルゴリズムに関する研究

    第28回情報理論とその応用シンポジウム予稿集 

  • 帰還通信路を用いた誤り制御方式に関しての研究

    第28回情報理論とその応用シンポジウム予稿集 

  • A Note on HTTP Traffic Analysis of the Time Series Model with a Time Varying Density Parameter

  • 使用ユーザが変化するDS/CDMAシステムにおけるベイズ最適なマルチユーザ検出について

    第28回情報理論とその応用シンポジウム予稿集 

  • 単一ループをもつグラフィカルモデルにおける確率伝播型アルゴリズムに関する一考察

    第28回情報理論とその応用シンポジウム予稿集 

  • 再帰的畳み込み符号を利用したReliability Based Hybrid ARQについての研究

    電子情報通信学会技術研究報告IT2005-52 

  • FSMX情報源に対するベイズ符号のメモリ量削減アルゴリズム

    電子情報通信学会技術報告IT2005-47 

  • 電子透かしにおける秘匿容量の計算手法に関する研究

    電子情報通信学会技術報告 IT2005-47 

  • A Study of Reliability Based Hybrid ARQ Scheme with Bitwise Posterior Probability Evaluation from Message Passing Algorithm

  • A Note on Construction of Nonlinear Unequal Orthogonal Arrays from Error-Correcting Codes

  • インターネットを用いた研究支援環境〜電子会議システム〜

    経営情報学会2005年春季全国研究発表大会予稿集 

  • 統計的決定理論に基づく有限受信バッファSR ARQ方式

    第28回情報理論とその応用シンポジウム予稿集 

  • ID情報に基づくランプ型非対称鍵配送方式について

    暗号と情報セキュリティシンポジウム2006予稿集 

  • 統計的決定理論に基づく複数のクラスに属する文書の分類方法に関する一考察

    電子情報通信学会技術報告IT2005-89 

  • 相互通信可能なネットワーク上での情報伝送に関する一研究

    第29回情報理論とその応用シンポジウム予稿集 

  • On Factorial Effects Corresponding to Orthogonal Arrays with Unequal Strength

    Proceeding of 2006 Hawaii, IEICE and SITA Joint Conference on Information Theory 

  • A Note on Transmission Schemes with Unequal Error Protection Codes and a Feedback Channel

    Proc. of Society of Information Theory and its Applications (SITA2006) 

  • 未知パラメータを伴う隠れマルコフモデルの状態推定に関する一考察

    第29回情報理論とその応用シンポジウム予稿集 

  • A note on the epsilon-overflow probability of lossless codes

  • Multiuser Detection Algorithms for CDMA based on the Massage Passing Algorithms

  • Adaptive Increment Redundancy Coding and Decoding Schemes using Feedback Information

  • A Note on Transmission Schemes with Unequal Error Protection Codes and a Feedback Channel

    Proc. of Society of Information Theory and its Applications (SITA2006) 

  • 秘密情報を持つBroadcast Channel の Secrecy Capacity 計算アルゴリズム

    第29回情報理論とその応用シンポジウム予稿集 

  • 記憶のある通信路に適した誤り訂正符号の構成法に関する研究

    第29回情報理論とその応用シンポジウム予稿集 

  • 記号の出現パターンを考慮した情報源に対するベイズ符号に関する研究

    電子情報通信学会技術研究報告IT2006-29 

  • On Factorial Effects Corresponding to Orthogonal Arrays with Unequal Strength

  • The Reliability based Hybrid ARQ Scheme with both the Encoded Parity Bit Retransmissions and Message Passing Decoding

  • 直交計画と双対符号の関係に関する一考察

    電子情報通信学会研究技術報告 

  • ベイズ符号化アルゴリズムを用いたテキストデータ圧縮

    情報処理学会研究報告 

  • 変動要因を考慮した非定常ポアソンモデルに関する一考察

    電子情報通信学会技術研究報告 

  • ランプ型Robust 秘密分散法に関する一考察

    暗号と情報セキュリティシンポジウム2007予稿集 

  • 画像に対する状態表現を用いたモデル化と無歪み符号化

    第30回情報理論とその応用シンポジウム予稿集 

  • 盗聴・改ざんに対して耐性を持つネットワーク符号化について

    第30回情報理論とその応用シンポジウム予稿集 

  • 外れ値データの発生を含む回帰モデルに対するベイズ予測アルゴリズム

    情報処理学会研究報告 

  • 混合情報源に対するε達成可能なオーバーフローしきい値に関する考察

    第30回情報理論とその応用シンポジウム予稿集 

  • 差分一様性を利用したMACの改良について

    暗号と情報セキュリティシンポジウム, SCIS 2007 

  • New Bounds for PMAC, TMAC, and XCBC

    Fast Software Encryption, 14th International Workshop, FSE 2007, Luxembourg, Luxembourg, March 26-28, 2007, Revised Selected Papers. Lecture Notes in Computer Science 4593 Springer 2007 

  • Tweakable Enciphering Schemes from Hash-Sum-Expansion

    Progress in Cryptology - INDOCRYPT 2007, 8th International Conference on Cryptology in India, Chennai, India, December 9-13, 2007, Proceedings. Lecture Notes in Computer Science 4859 Springer 2007 

  • A Note on Morphological Analysis Methods based on Statistical Decision Theory

    SICE 2007 PROCEEDINGS 

  • スペルミスを伴う形態素解析に関する一考察

    第30回情報理論とその応用シンポジウム予稿集 

  • On the Condition of ε-achievable Overflow Thresholds for Mixed Sources

  • マーキング仮定に基づくフィンガープリンティング符号のキャパシティについて

    暗号と情報セキュリティシンポジウム2008(SCIS 2008) 

  • Efficient Domain Extension Using Weak Pseudorandom Function

    暗号と情報セキュリティシンポジウム, SCIS 2008 

  • A Note on the Weak Universal Joint Source-Channel Coding

  • 複数のLDPC符号の交錯による有限状態マルコフ通信路に適した誤り訂正符号の構成法について

    電子情報通信学会技術研究報告 

  • An Efficient Bayes Coding Algorithm using a New Unlimited Depth Context Tree

  • 実験計画法における効果の推定の計算量削減に関する一考察

    第31回情報理論とその応用シンポジウム予稿集 

  • A Note on Automatic Construction Algorithms for Orthogonal Designs of Experiments Using Error-Correcting Codes

    Pre-ICM International Convention on Mathematical Sciences 

  • A Note on Automatic Construction Algorithms for Orthogonal Designs of Experiments Using Error-Correcting Codes

    Pre-ICM International Convention on Mathematical Sciences 

  • An Accurate Density Evolution Analysis for a Finite-State Markov Channel

    Proc. of Society of Information Theory and its Applications (SITA2008) 

  • 区間で一定なパラメータを持つ非定常情報源の漸近的な性質について

    第31回情報理論とその応用シンポジウム予稿集 

  • ブロック誤り率が未知の場合の選択再送ARQに関する一考察

    第31回情報理論とその応用シンポジウム予稿集 

  • On the Overflow Probability of Lossless Codes for Mixed Sources

  • A Note on the Iterative Interference Cancellation and Decoding for Coded CDMA

  • A Linear Programming Bound for Unequal Error Protection Codes

  • 分枝カット法に基づいた線形符号の復号法に関する一考察

    第32回情報理論とその応用シンポジウム予稿集 

  • A Note on the Second Order Separate Source Coding Theorem for Sources with Side Information

  • Document Classification Method with Small Training Data

    ICCAS-SICE2009 

  • A Note on Theoretical Limit of Type-I Hybrid Selective-repeat ARQ with Finite Receiver Buffer

    ISCTA'09 

  • 通信路状態が未知の選択再送ARQに関する一考察

    第32回情報理論とその応用シンポジウム 

  • A Note on Fixed-Length Coding Theorem for Sources with Side Information

  • 相関のある情報源における符号化定理に関する一考察

    第7回シャノン理論ワークショップ予稿集 

  • A Note on Estimation of the Effects in the Experimental Design using Fourier Transforms

  • サービスの開始と終了を考慮したWebトラヒックの非定常Poisson過程によるモデル化について

    電子情報通信学会技術研究報告 

  • Applying Markov Decision Processes to Selective-repeat ARQ with Finite Receiver Buffer

    ISMAC2010 

  • Probabilistic World Logic for Modality of Uncertainty

▼display all

Specific Research

  • 情報理論的完全情報保護のもと効率的かつ信頼性が高い分散計算システムの理論基盤構築

    2020  

     View Summary

    本研究の目的は,情報理論的に完全に情報を保護したもとでの高信頼性を有する高効率な分散計算システムの理論基盤を構築することである.本研究では,効率性,安全性,秘匿性を考慮した分散処理システムを数理的に定式化し,各評価基準の最適値のトレードオフ解析を行い,最適性を達成する分散計算システムを構築する手法を採用した.成果の一つとして,情報理論的に完全にプライバシ保護したもとでの通信の効率が最適である情報検索システムと,サーバの情報が他サーバの情報から最適な効率で復元可能な分散ストレージシステムの対応関係を,集合等の記法を用いて簡明に整理したことが挙げられる.本成果は3月の情報理論研究会にて公開された.

  • 高信頼性, 高安全性を有する高速な分散処理システムの研究

    2019  

     View Summary

    近年,個人情報を含むビッグデータが増え,そのようなデータを処理するシステムが求められている.その際,データを盗聴者から安全に保護しつつ処理するという安全性,高速に処理を行うという計算処理時間,処理の際に誤りが起こっても正確に処理できるという信頼性,という3つの評価基準が重要になる.従来は,データを分散してから処理することにより,いずれかの評価基準において良くなるようなシステムが研究されてきた.例えば,データを複数のサーバに分散し並列処理を行うことで高速に信頼性の高い処理を行うというcoded computationや,データを複数のストレージに分散することで安全に保護するという分散ストレージがある.これらは,数理モデルを定式化することで厳密に議論されてきた.しかし,すべての評価基準を考慮に入れたシステムに関しては,現在十分な研究があるとは言えない.本研究は,coded computation に分散ストレージに関する理論を取り入れることで,高信頼,高安全性を有する高速な分散処理システムの理論を確立を目指した.具体的には,単純に組み合わせるのではなくシステム全体について厳密に議論するため,システム全体の数理モデルを定式化し,性能の評価基準を明確に定義し,その下で性能の理論限界を導出し,それを達成するシステムを構築した.

  • ネットワーク知能の基礎数理モデルに向けた情報理論の拡張

    2017  

     View Summary

    ネットワーク知能とも呼ぶべき近年の人工知能における,推論,学習の最適性や理論限界を明確にするための基礎理論構築に向けた情報理論の拡張および,ネットワーク上で相互に通信を行いながら所望の推論結果を得るための通信アルゴリズムに関する研究を行った.情報理論の拡張としては特に,歪みのある情報源符号化における理論限界の導出を行った.また,可変長intrinsic randomness問題における様々な理論限界の導出および,それらと他の情報理論の問題における理論限界との対応関係を明らかにした.通信アルゴリズムに関する研究としては特に,ランク誤りを考慮したcoded computationのための通信アルゴリズムを提案した.

  • ネットワークデータ処理のためのスパースモデリングの拡張と数理基盤の構築

    2016  

     View Summary

    スパースモデリングによるビッグデータ解析の研究は多くの成果を収めているが,その範囲はデータの解析部分がほとんどで,解析に至るまでの収集,蓄積,検索,通信等のネットワークデータの処理プロセス全体についてはほとんど考察されていない.また,解析を中心とした研究では評価基準として解析精度のみに注目した考察で十分であったが,通信量,蓄積量等の効率の評価基準や情報保護,漏洩情報量等の安全性の評価基準を同時に考察することが必要となる.そこで,本研究では,ネットワークデータの収集,分散蓄積,分散解析の3つの処理プロセスに対して,数理モデルによる問題のモデル化と評価基準の数理的定義を行ない,問題の定式化と解くべき問題を整理した.その後,それらの問題に対して最適なあるいは近似最適なアルゴリズムの構成やその性質の理論的考察を行なった.さらに,アルゴリズムの実装と実験による性能評価を行なった.

  • ベイズ符号による無歪み画像圧縮に関する研究

    2013  

     View Summary

     従来の無歪画像圧縮に関する研究は,アドホックな方法によるアルゴリズムの開発と,試行錯誤を繰り返すことによるアルゴリズムの改良というアプローチを採っており、理論的最適性などが保証されていることはあまりない.これらの研究は,この根本的問題を解決しない限り,画像圧縮の研究が大きな進展を得ることは期待できない.これに対し,本研究ではまず,画像に対して適した確率モデルを構築し,その確率モデル上で最適性を保証するベイズ符号を構成するアプローチをとることで,上記の根本的問題の解決を試みるものである. まず,本年は研究目的のため,画像データには特化せず,より基本的な情報源を考え,ベイズ符号やベイズ符号を用いた予測についての理論的な解析と画像データの確率モデルの基礎となるいくつかの確率モデルに対して,最適または近似最適な効率的なアルゴリズムの構成とその性能評価をおこなった.(1) 真のモデルとベイズ符号が仮定したモデルの符号長における影響の漸近解析 ベイズ符号は,パラメタライズされた分布を仮定しその確率パラメータを未知としてベイズ基準で符号長最小の符号化となっているが,もし仮定した分布で真の分布が表現できない場合,ベイズ基準の意味での最適性は保証されない.しかし,真の分布と仮定した分布の中でK-L情報量的に最も近い分布とのK-L情報量だけ漸近符号長が長くなるものの,ベイズ符号が優れた性質もつことを証明した. 特に,画像データのように階層的な確率モデルが仮定される場合に,真の画像データの分布がその仮定した階層確率モデル族で表現できない場合でも,階層モデルの中で,平均符号長を最小とするモデルに漸近収束することが示され,真の分布を,仮定した確率モデルが含まない場合でも,ベイズ符号が優れた性質を持っていることを示した.(2) ベイズ符号のオーバーフロー確率の理論評価 ベイズ符号が平均符号長を評価基準としてベイズ基準,minimax基準,maxmin基準において漸近的に最適な符号であることはよく知られているが,オーバーフロー確率の面での評価は今までに行なわれていなかった.実用的な見地からは,平均符号長における評価のみならず,符号長が,あるメモリー量をオーバーしてしまう確率であるオーバーフロー確率も重要な評価指標である.そこで,符号長の2次モーメントや自己エントロピーの2次モーメントを用い,オーバーフロー確率の漸近解析を行なった.その結果この面においてもベイズ符号が優れた特性を有することが示された.(3) 文脈木モデル以外の拡張された情報源に対するベイズ符号アルゴリズムの構築 情報源符号化の研究の多くは,情報源の確率的な性質は時間によって変化しないとする定常情報源を対象としている.文脈木モデルも例外ではない.ところが,実データにおいては,幾つかの性質が異なるコンテンツが1つのファイルに含まれていることも多く,そのようなファイルを情報源として見た場合には,非定常な情報源と考えるのが自然である.情報源が非定常であるにも関わらず,定常情報源の場合と同様の符号化アルゴリズムを使用すれば,良い圧縮性能を望むことは出来ない.情報源が非定常であれば,符号化アルゴリズムもその変化を追従するアルゴリズムを構築する必要がある.一言に非定常情報源と言っても,モデル化の方法は一意ではなく,幾つかのモデルが考えられる.本研究グループは,これらのモデルの内の一部である区間で定常な非定常情報源やパラメータが時間によって変化する文脈木情報源に対するベイズ符号を提案し,性能に関する理論解析を行い,さらにベイズ符号を実現する効率的アルゴリズムを提案した.  また,実データに対する符号化を考えると,データのサイズが小さい場合には,全種類の情報源アルファベットが出てこない場合が考えられる.このような場合に,従来と同様の符号化をしてしまうと,圧縮の性能が劣化してしまうことが知られている.本研究グループは,情報源アルファベットの出現パターンを考慮した情報源モデルに対するベイズ符号を提案し,性能に理論解析を行い,さらにベイズ符号を実現する効率的なアルゴリズムを提案した. それらの研究の拡張として,画像データや言語データ等にも適した非定常な情報源モデルや情報源アルファベットが部分的に出現する情報源モデルの検討を行い,ベイズ符号の性能に関する理論解析及び,ベイズ符号を実現する効率的アルゴリズムの構築を行った.

  • 確率推論アルゴリズムに基づいたストリーム暗号への統一的攻撃法安全性に関する研究

    2009  

     View Summary

    従来のストリーム暗号の研究では,非線形コンバイナ型乱数生成器の高速攻撃法の一部で通信路符号化の高効率復号アルゴリズムが応用されていたが、攻撃を確率推論の問題としてモデル化したもとで攻撃法を解析した研究は存在しなかった。また、暗号の安全性研究の多くは、それぞれの暗号に対して個別の攻撃法の提案が行なわれ、それに基づき安全性評価が行なわれてきた。しかしこのようなアプローチでは、より強力な攻撃法が新たに提案される可能性が残り、安全性の保証が非常に不安定であった。本研究では、まず、非線形コンバイナ型のみならず全てのストリーム暗号に対する攻撃を確率推論の問題として定式化し統一的な枠組を構築した。次に、この統一的枠組のもとで、最適な攻撃法をベイズ決定理論から理論的に導出した。この最適解が明示されたことにより、非線形コンバイナ型のみならず、従来研究では効率的攻撃法が見いだせなかった様々な種類のストリーム暗号に対しても攻撃法を考察できるようになった。この最適解を求める(最適な攻撃を行なう)ためには、一般に乱数生成器の鍵の長さの指数オーダの計算量が必要であるため、効率的な確率推論アルゴリズムを適用することで莫大な計算量を削減する必要がある。そのため、本質的に同等な確率推論問題を扱っている周辺分野で、高い性能を有していることが実証されているアルゴリズム、例えば、符号理論や統計力学の分野で多く用いられている和積(sum-product)アルゴリズム等の応用を考察した。また、信号処理や実験計画の視点からのアルゴリズムについてもいくつか考察を行なった。その他として、基盤となるベイズ決定理論についても、様々な応用分野について広く考察し、最適または近似的に最適なアルゴリズムを導出することにより、ストリーム暗号への応用の基礎がためを行なった。さらに、乱数生成自体やそれと同質の問題である情報源符号化の問題に対しても、情報理論的視点からその限界に対する考察も行なった。

  • 大規模分散ネットワーク情報の処理に関する基礎モデルの構築と特性解析

    2006  

     View Summary

     ネットワーク上で分散して存在する膨大な情報をいかに処理するかは,今後ますます重要な問題となってくると考えられる.しかし,この分野では数理的基礎モデルが十分でなく,最適解や性能限界について議論がなされていないため,今後のこの分野における発展のための道標は明確とは言えない. そこで本研究では,ネットワーク分散情報の処理における,実システムを視野に入れた,この分野における基礎理論の構築を目指す.数理的基礎モデルの構築から実際のアルゴリズム・システム設計までの段階的研究アプローチの一環として,まず基礎モデルの構築と,最適解や理論的性能限界についての定式化を目的としていた.大きくわけて以下の2つのステップに分けて研究を行った.①ネットワーク上に分散した情報の伝送と決定・制御に関する多端子情報理論と統計的決定理論を用いた基礎モデルを構築し,構築したモデル上で各問題の目的に応じた評価関数を定義し,最適解や理論的性能限界についての定式化を行う.②定式化した最適解またはその近似解を実現するアルゴリズムをグラフ上のメッセージ伝播アルゴリズム等を用いて設計し,その性能や計算量の評価を解析的・実験的に行う.

  • メディアネットワークにおける高圧縮符号化技術に関する研究

    2002  

     View Summary

    多種多様な形態の情報が多量に伝送,蓄積される,ネットワーク情報化社会において,情報の圧縮技術は欠く事のできない基盤技術となりつつある.情報圧縮の符号化(情報源符号化)の領域は大別して次の2つがある.①無歪圧縮(元の情報系列に可逆的に戻るもの)②有歪圧縮(元の情報系列に可逆的に戻らないもの)本研究では主に①の無歪圧縮,特に無歪ユニバーサル符号化について扱った.情報源の確率構造が既知の場合はShannon-Fanoの符号化,Huffmanの符号化等を用いることにより理論的圧縮限界であるエントロピーまで圧縮可能なことはよく知られている.しかし,現実のテキスト,画像の情報系列はその情報源の確率構造が未知である場合がほとんどで,このような場合が無歪みユニバーサル符号化の問題である.現在無歪ユニバーサル符号化アルゴリズムとして広く実用化されているZiv-Lempel符号は定常エルゴード情報源に対し漸近的にエントロピーまで圧縮できる優れた符号であるが,その収束速度は遅く,最適でない事が知られている.そこに近年,無歪みユニバーサル符号の収束の速さの理論的限界がアメリカの研究者により明らかになると共にそれを達成する符号(例えばベイズ符号)が提案された.また,木構造を有効に用いた符号化のアルゴリズム(CTW法)もオランダの研究グループにより提案され,理論ばかりでなく実用化についても可能性が開けつつある.本研究グループも理論と符号化アルゴリズム両面でアメリカ,オランダの研究者たちとほとんど同時期に同様の成果を発表し,その後もこの研究を発展させている.しかし,これらのアルゴリズムは,計算量・メモリー量の面で問題があり,まだ実用的ではなかった.そこで,本研究では無歪みユニバーサル符号の計算量・メモリ量において実用的なアルゴリズムの構築を行い,理論的,実験的にその性能の解析および評価を行った.また,情報源が時系列的に変化する場合のユニバーサル符号の提案を行い,その最適性の証明及び実験による評価も行った.さらにこの研究の拡張として,このベイズ的に最適なユニバーサル符号を信号処理や分散協調処理の問題へ応用しいくつかの成果が得られた.

  • 情報理論に基づく学習の一般モデル・アルゴリズムの研究

    2000  

     View Summary

     本研究は学習を様々な研究領域について横断的に、情報の視点から考察していく。特に、情報の視点では情報理論と決定理論から学習における情報の流れを整理し、学習の目的や学習対象から学習者にどのように情報が与えれられるかのモデルについて考察する。 学習問題は情報量の観点から簡単に説明がつき、対象から得られた情報から学習目的のために必要な情報を最大限抽出することが最適な学習法と位置づけられる。この最適な学習は情報理論のユニヴァーサル情報源符号化の限界やベイズ決定理論における最悪の事前分布の設定問題と密接な関係にあることも明らかになってきている。 本研究では上記の今までの成果をふまえて、人工知能の機械学習の問題のみならず、さらに様々な領域の学習問題に対して情報理論や決定理論などを基礎とする情報の視点から横断的に考察を行っていくことになり、これまでに以下のような結果を得ている。 各学習問題の目的、学習結果の評価基準、学習対象から与えられる情報の流れのモデル、学習アルゴリズム等を情報の視点で再整理し、その問題個別の特徴、全てに共通の性質などを明確にした。 情報の視点から本質的には同等のアルゴリズムを抽出し、それを一般化することで学習の基本モデルとアルゴリズムを新たに構築した。このアルゴリズムの限界や性質について情報理論や決定理論をはじめ幾つかの視点から考察し、また、それぞれの個別の問題で有効なアルゴリズムについては、学習問題の条件から、そのアルゴリズムの有効性に効いている条件を情報の視点から分類整理し相互の関連を明らかにした。 さらに、各分野の個々の学習問題に対して構築した基本モデルとアルゴリズムの適用を試み、さらにその問題の特殊性が分類整理された学習モデルの条件のどれと類似しているかを検討し、有効なアルゴリズムのそのままの適応やそれを拡張した新たなアルゴリズムの提案を行った。個々の問題において適用あるいは提案したアルゴリズムの性質についても理論面からの考察と、シミュレーションなど実験面からの考察も行った。

  • メディアネットワークにおける高圧縮符号化技術に関する研究

    1998  

     View Summary

    ネットワーク情報化社会において、情報の圧縮技術は欠く事のできない基盤技術となりつつある。情報圧縮符号化の領域は大別してテキストデータ等の圧縮に利用されている無歪(可逆)圧縮と音声や画像の圧縮に利用されている有歪(非可逆)圧縮の2つがある。 前者では Ziv-Lempel符号を用いた符号が実用化されており、この符号は漸近的に理論的圧縮限界であるエントロピーまで圧縮可能な優れた符号である。しかし、その収束の速さは遅く、最適でないことが知られている。1%でも圧縮率を上げたいという情報化時代のニーズから考えると、新たな符号の提案、実用化が待たれていたが、近年、情報理論の分野でユニヴァーサル無歪圧縮の収束の速さの理論的限界とそれを達成するベイズ符号が提案された。また、木構造を有効に用いたこの符号のアルゴリズムも提案され、理論ばかりでなく実用化についても可能性が開けつつある。研究者たちも理論と符号化アルゴリズム両面でほとんど同時期に同様の成果を発表し、その後もこの研究を発展させ、理論と応用の両面で幾つかの成果を得た。 また、このベイズ的アプローチは統計的推論、モデル選択や学習理論の面への応用も可能で、それらの分野においても新たなアルゴリズム提案や性能の漸近的評価等幾つかの成果を得ることが出来た。

  • マルチメディアにおける高圧縮符号化技術に関する研究

    1997  

     View Summary

    ネットワーク情報化社会において、情報の圧縮技術は欠く事のできない基盤技術となりつつある。テキストデータ等の圧縮に利用されている無歪(可逆)圧縮では、圧縮限界であるエントロピーへの収束の速さの理論的限界を達成するベイズ符号が近年提案された。また、木構造を有効に用いたこの符号のアルゴリズムも提案され、理論ばかりでなく実用化についても可能性が開けつつある。研究者たちも理論と符号化アルゴリズム両面でほとんど同時期に同様の成果を発表し、その後もこの研究を発展させている。 本研究は無歪みのユニバーサル符号であるベイズ符号の基礎研究が起点となる。ベイズ符号の限界や性質などの情報理論的研究はベイズ決定理論からの理論的解析が中心となる。またベイズ符号の実用化に向けての計算機科学的研究はアルゴリズム論による考察や計算機実験による性能評価が中心となる。この理論、実用両方向からの研究を独立して行うのではなく交互のやりとりにより幾つかの知見が得られた。 理論研究では、より広い情報源のクラスやメモリーが有限の場合の圧縮限界について研究を行なった。実用化アルゴリズムの検討としては、シミュレーションにより、上記の理論式との対応を検討すると共にいくつかの既存のアルゴリズムとの融合も試みた。例えば、ベイズ符号についての理論的共通点が指摘されており、実用化のためのさまざまな工夫が既に行われているZiv-Lempel符号の実用アルゴリズムの一部が、ベイズ符号にも流用可能であれると考えられ、本研究で新たに提案した両者の中間的なアルゴリズムが Ziv-Lempel符号の定数倍の計算量で、FSMX情報源に対してZiv-Lempelより低い圧縮率を達成した。また、ベイズ符号やベイズ最適決定を、モデル選択や知識情報処理などに応用した研究も行い、幾つかの成果が得られている。研究の成果発表:1997/6, IEEE Int. Symp. on IT, Asymptotic property of sufficient statistic codes 1997/6, IEEE Int. Symp. on IT, A study on difference of codelengths between MDL codes and Bayes codes on case different priors are assumed1997/9, 電子情報通信学会論文誌A, Berlekamp-Masseyアルゴリズムを用いたBCH限界を超える復号法の計算量について1997/10 IEEE Int. Symp. on SMC, A learning with membership queries to minimize prediction error1997/10 IEEE Int. Symp. on SMC, Machine learning by a subset of hypotheses1997/10 IEEE Int. Symp. on SMC, A new architecture of signature analyzer for multi-output circuit1998/3, 人工知能学会誌, 矛盾を含む知識の取り扱いについての一考察1997/5, 信学技報/電子情報通信学会, ハッシュ技法によるデータ探索の数学的モデル化,及び探索効率の漸近的評価1997/5, 春季大会予稿集/日本経営工学会, 属性値の類似度を用いた概念学習の効率1997/5, 経営情報学会, 属性のクラスタを用いた概念学習の効率化1997/6, 人工知能学会全国大会(第11回)論文集/人工知能学会, 統計的決定理論によるデータ検索の定式化と最適化1997/7, 信学技報/電子情報通信学会, ベイズ決定理論に基づく統計的モデル選択について1997/7, 信学技報/電子情報通信学会, ベイズ統計学に基づく計算論的学習モデルと学習可能性1997/7, 信学技報/電子情報通信学会, 構造モデル族の学習,予測アルゴリズムに関する一1997/7, 信学技報/電子情報通信学会, トレリス符号を用いた有歪みデータ圧縮の一考察1997/7, 信学技報/電子情報通信学会, 不均一誤り訂正符号の復号法に関する一考察1997/12, 第20回情報理論とその応用シンポジウム予稿集, 階層的確率モデルにおけるタイプについて1997/12, 第20回情報理論とその応用シンポジウム予稿集, 不確実性を含むデータの統合に関する一考1997/12, 第20回情報理論とその応用シンポジウム予稿集, ベイズ符号の視点からのZiv-Lempel78符号の改良に関する一考察1997/12, 第20回情報理論とその応用シンポジウム予稿集, ベイズ符号化法におけるメモリの問題点に関する一考察1997/12, 第20回情報理論とその応用シンポジウム予稿集, 決定を考慮したベクトル量子化法の提案1997/12, 第20回情報理論とその応用シンポジウム予稿集, 木構造型モデル族のモデル選択法に関する一考察1997/12, 第20回情報理論とその応用シンポジウム予稿集, 混合分布の近似とその性能について1997/12, 第20回情報理論とその応用シンポジウム予稿集, 直交表現された仮説の学習に関する一考察1997/12, 第20回情報理論とその応用シンポジウム予稿集, 質問からの学習における予測誤りに関する一考察1997/12, 第20回情報理論とその応用シンポジウム予稿集, BCH限界を超える復号アルゴリズムを用いたChase復号法の計算量低減1998/1, 信学技報/電子情報通信学会, 階層モデル族のモデル選択における選択誤り率について

  • 情報理論に基づく柔らかな情報処理に関する研究

    1995  

     View Summary

    人間のもつ学習・認識などの機能をマシン上に実現し,あいまいや誤った情報にも適応できる柔らかい情報システム実現のためには従来の理論を基礎とした体系では不十分であり,論理プラスアルファの体系が必要となってきている。 研究者らは柔らかい情報処理の要素である不確実な推論と学習のための基礎理論体系として,第一階言語の解釈空間に確率測度を導入した確率世界論理を数年前より提案し,いくつかの推論を情報の変換としてこの基礎体系上でモデル化し,情報理論の立場から評価基準の明確化や最適な推論アルゴリズムの提案を行っている。 今年度得られた成果としては,矛盾を含んだ知識の管理法と推論法,配列構造を用いた不確実な知識の表現と推論法,質問を許す学習の最適化,収益を最大化を目的とする強化学習アルゴリズムなどが上げられる。 例えば,矛盾を含んだ知識の管理法と推論法では,矛盾が生じるメカニズムを情報理論のモデルにより定式化することにより新たな手法を提案している。質問を許す学習の最適化では,質問回数一定のもとで誤り率最小や誤り率一定のもとで平均質問回数最小の基準の下で最適な学習アルゴリズムを情報理論的上界をうまく使うことによって,提案している。

▼display all

Overseas Activities

  • 多端子情報理論の基礎となる新たな基礎数理の構築

    2011.04
    -
    2012.03

    アメリカ   University of California Berkeley

  • ベイズアプローチによる学習理論と事後確率の効率的計算アルゴリズム

    2001.10
    -
    2002.07

    アメリカ   ハワイ大学

 

Syllabus

▼display all