研究業績(論文)

2017

  1. Joe Suzuki, “E-learning Design and Development for Data Science in Osaka University”,  Conference on Education of Data Science, Shiga University, Japan, November 2017
    [Conference Paper, invited] (slide)
  2. Joe Suzuki, “Conditional Mutual Information Estimation and its application to Conditional Independence Detection“, The Tenth Workshop on Information Theoretic Methods in Science and Engineering (WITMSE) , Paris, France, September 2017.
    [Conference Paper, invited] (slide)
  3. Joe Suzuki and Jun Kawahara, “Branch and Bound for Regular Bayesian Network Structure learning“, Uncertainty in Artificial Intelligence, pages 212-221,  Sydney, Australia, August 2017.
    [Conference Paper] (slide)
  4. Joe Suzuki, “Klein’s fundamental 2-form of second kind for the Cab curves“,  Symmetry Integrability and Geometry Methods and Applications 13(17), 13 pages,  March 2017.
    [Journal Paper]
  5. Joe Suzuki and Jun Kawahara, “BNSL“, CRAN R Package, March 2017.
    [Software]
  6. Joe Suzuki, “A Theoretical Analysis of the BDeu Scores in Bayesian Network Structure Learning“, Behaviormetrika 1(1):1-20 · January 2017 (online version: November 2016).
    [Journal Paper]
  7. Joe Suzuki, “An Efficient Bayesian Network Structure Learning Strategy“, New Generation Computing 21(1), January 2017 (online version December 2016)
    [Journal Paper]

2016

  1. Joe Suzuki, “A novel Chow–Liu algorithm and its application to gene differential analysis“, International Journal of Approximate Reasoning 80, January 2017 (online version August 2016)
    [Journal Paper]
  2. Joe Suzuki, “Jeffreys’ and BDeu Priors for Model Selection“, The Ninth Workshop on Information Theoretic Methods in Science and Engineering (WITMSE) ,  Helsinki, Finland, September 2016.
    [Conference Paper, invited]
  3. Joe Suzuki, “Structure learning and universal coding when missing values exist“,  IEEE International Symposium on Information Theory (ISIT) 2016, Barcelona, Spain,  July 2016
    [Conference Paper] (slide)
  4. Joe Suzuki,  “Miura“,  MaCaulay 2 Package,  April 2016
    [Software]
  5. Joe Suzuki, “An Estimator of Mutual Information and its Application to Independence Testing“, Entropy 18(4):109, March 2016
    [Journal Paper]
  6. Joe Suzuki, “Structure Learning of Bayesian Networks with p Nodes from n Samples when n<<p
    “, Probabilistic Graphical Model Workshop: Sparsity, Structure and High-dimensionality, Tokyo, Japan.  March 2016.
    [Conference Presentation] (slide)

2015

  1. J. Suzuki. “Efficiently learning Bayesian network structures based on the B&B strategy: A theoretical analysis“, Advanced Methodologies for Bayesian Networks, Japanese Society of Artificial Intelligence, Yokohama, Japan, Nov. 2015  (Lecture Note on Artificial Intelligence 9505, pp. 1-14, , Springer-Verlag).
    [Conference Paper] (slide)
  2. J. Suzuki, “Forest learning based on the Chow-Liu algorithm and its application to genome differential analysis: A novel mutual information estimation“,  Advanced Methodologies for Bayesian Networks, Yokohama, Japan, Japanese Society of Artificial Intelligence,  Nov. 2015 (Lecture Note on Artificial Intelligence 9505, pp. 234-249,  Springer-Verlag).
    [Conference Paper] (slide)
  3. J. Suzuki, “Consistency of learning Bayesian network structures with continuous variables: An information theoretic approach“, Entropy 17(8): 5752-5770, August 2015.
    [Journal paper]

2014

  1. T. Ayano and J. Suzuki. “Bayes independence test”. Workshop on Graph-based Algorithms for Big Data and its Applications, Yokohama, Japan, Nov. 2014
    [Conference Presentation] (slide)
  2. J. Suzuki. “The Chow-Liu algorithm based on the MDL when discrete and continuous variables are present”.  The 4th Workshop on Algorithmic issues for Inference in Graphical Models, Paris, France, September 2014.
    [Conference Presentation] (slide)
  3. J. Suzuki. “Learning Bayesian network structures when discrete and continuous variables are present“. The sixth European workshop on Probabilistic Graphical Models, Utrecht, Netherlands, Sept. 2014 (Lecture Note on Artificial Intelligence 8754: 471-486,  Springer-Verlag).
    [Conference Paper] (slide)

2013

  1. J. Suzuki. “The universal Bayesian Chow-Liu algorithm”. Workshop on Data Discretization and Segmentation for Knowledge Discovery, Japanese Society of Artificial IntelligenceYokohama, Japan, Oct. 2013.
    [Conference Presentation] (slide)
  2. Joe Suzuki, “The MDL principle for arbitrary data:either discrete or continuous or none of them“,
    The Sixth Workshop on Information Theoretic Methods in Science and Engineering (WITMSE),
    Tokyo, Japan, August 2013
    [Conference Paper, invited] (slide)
  3. J. Suzuki. “Universal Bayesian measures“. IEEE International Symposium on Information Theory pp. 644-648, Istanbul, Turkey, July 2013.
    [Conference Paper] (slide)
  4. J. Suzuki. “Network coding and polymatroid (simple survey)”. Asia and Europe Workshop on Information Theory, Kamakura, Japan, May 2013.
    [Conference Presentation] (slide)

2012

  1. T. Ayano and J. Suzuki. “On d-asymptotics for high-dimensional discriminant analysis with different variance-covariance matrices“,  IEICE trans. on Information and Systems E95-D(12): 3106-3108,  December 2012.
    [Journal Paper]
  2. J. Suzuki, “The Hannan-Quinn proposition for linear regression“. International Journal of Statistics and Probability 1(2), Nov. 2012.
    [Journal Paper]
  3. J. Suzuki. “Universal prediction without assuming either discrete or continuous”. The Fourth Workshop on Game-Theoretic Probability and Related Topics, Tokyo, Japan, Nov. 2012.[Conference Presentation]  (slide)
  4. J. Suzuki. “Bayesian criteria based on universal measures“. International  Symposium on Information Theory and its Applications: 71-75, Honolulu, Hawaii, Oct. 2012.
    [Conference Paper]  (slide)
  5. J. Suzuki. “The Bayesian Chow-Liu algorithms“. The sixth European workshop on Probabilistic Graphical Models, pp. 315-322, Granada, Spain, Sept. 2012.
    [Conference Paper]  (slide)
  6. J. Suzuki. “Bayesian network structure learning for discrete and continuous variables“. IEEE International Conference on Uncertainty and Knowledge Engineering, pp. 141-144, Jakarta, Indonesia, August 2012
    [Conference Paper] (slide)
  7. J. Suzuki. “Bayesian network structure estimation based on the Bayesian/MDL criteria when both discrete and continuous variables are present“. IEEE Data Compression Conference, pp. 307-316, Snowbird, Utah, April 2012.
    [Conference Paper] (slide)

2011

  1. J. Suzuki. “MDL/Bayesian criteria based on universal coding/measure“, Solomonoff 85th Memorial Conference, Melborne, November 2011 (Lecture Notes on Computer Science 7070: 399-410,  Springer Verlag).
    [Conference Paper] (slide)
  2. T. Inazumi, T. Washio, S. Shimizu, J. Suzuki, A. Yamamoto, and Y. Kawahara. “Discovering causal structures in binary exclusive-or skew acyclic models“. In the 27th Conf. on Uncertainty in Arti cial Intelligence, pp. 373-382, Barcelona, Spain, July 2011.
    [Conference Paper] (slide)
  3. J. Suzuki. “The universal measure for general sources and its application MDL/Bayesian criteria”. Data Compression Conference, p. 478, Snowbird, Utah, March 2011. IEEE.
    [Conference Presentation] (slide)

2010

  1. J. Suzuki. “A Markov chain analysis on genetic algorithms: Large deviation principle approach“, Journal of Applied Probability 47(4): 967-975, December 2010.
    [Journal paper]
  2. J. Suzuki. “A generalization of nonparametric estimation and on-line prediction for stationary ergodic sources”. Asia-Europe Information Theory Workshop, pp. 17-18, Ishigaki Island, Oct. 2010.
    [Conference Presentation] (slide)
  3. J. Suzuki. “A generalization of the Chow-Liu algorithm and its applications to artificial intelligence”. International Conference on Artificial Intelligence, pp. 478-484, Las Vegas, Nevada, July 2010.
    [Conference Paper] (slide)

2009

  1. J. Suzuki. “A conjecture on strong consistent learning”. Workshop on Learning with Logics and Logics for Learning, Japanese Society of Arti cial Intelligence, Kyoto, Japan, July 2009.
    [Conference Presentation] (slide)

2007

  1. J. Suzuki. “Miura conjecture on affine curves“, Osaka Journal of Mathematics 44(1): 187-196, March 2007.
    [Journal paper]

2006

  1. J. Suzuki. “On strong consistency of model selection in classi cation“,  IEEE Trans. on Information Theory 52(11): 4767-4774, Nov 2006.
    [Journal paper]

2005

  1. J. Suzuki. “Strong consistency in learning stochastic relations”. Workshop on Learning with Logics and Logics for Learning, Japanese Society of Arti cial Intelligence, Kokura, Japan, June 2005.
    [Conference Presentation]
  2. C. DeSilva and J. Suzuki. “On the stationary distribution of GAs with positive crossover probability”. Genetic and Evolutionary Computation Conference, pp. 1147-1151, Washington DC, June 2005.
    [Conference Paper]

2004

  1. J. Suzuki. “Combinatorial source coding with costs“. IEEE Trans. on Information Theory 50(5):  777-780, May 2004.
    [Journal paper]

2003

  1. J. Suzuki. “Coding combinatorial sources with costs”. IEEE International Symposium on Information Theory, p. 113, Yokohama, Japan, June 2003.
    [Conference Paper]

2002

  1. [Journal Paper (in Japanese)]

2001

  1.  K. Hojo, B. Ryabko, and J. Suzuki. “Performance of data compression in terms of Hausdorff dimension”. IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E84-A(7): 1761-1764, July 2001.
    [Journal paper]
  2. B. Ryabko and J. Suzuki, “Comparing the Multilevel Pattern Matching Code and the Lempel-Ziv Codes”,  IEEE International Symposium on Information Theory,  Washington DC, June 2001.
    [Conference Presentation]
  3. R. Harasawa and J. Suzuki. “Jacobian group arithmetic for cryptography”. IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E84-A(1):130-139, Jan. 2001.
    [Journal paper]

2000

  1. .J. Suzuki and B. Ryabko, “Combinatorial Source Coding with Costs”,  International Symposium on Information Theory and its Applications, pages 77-81, Honolulu, Hawaii, November 2000.[Conference Paper]
  2. R. Harasawa and J. Suzuki, “Fast Jacobian Group Arithmetic on Cab Curves”, Lecture Note on Computer Science, the 4th Algorithmic Number Theory Sympojium, pages 359-376, July 2000.
    [Conference Paper]
  3. J. Shikata, Y. Zheng, J. Suzuki, and H. Imai. “Realizing the Menezes-Okamoto-Vanstone (mov) reduction efficiently for ordinary elliptic curves“. IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science: E83-A(4): 756-763, April 2000.
    [Journal paper]

1999

  1. J. Shikata, Y. Zheng, Joe Suzuki, and Hideki Imai, “Optimizing the Menezes-Okamoto-Vanstone Algorithm for Non-Supersingular Elliptic Curves”, Lecture Note on Computer Science 1716, Advances in Cryptology-Asiacrypt, November 1999.
    [Conference Paper]
  2. J. Suzuki. “Learning Bayesian belief networks based on the minimum description length principle: Basic properties“. IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E82-A (10): 2237-2245, Oct. 1999.
    [Journal paper]
  3. [Journal Paper (in Japanese)]
  4. Boris Ryabko, Joe Suzuki, Flemming Topsoe, “Hausdorff Dimension as a New Dimension in Source Coding and Predicion”, IEEE Information Theory workshop, Johannesburg, South Africa, June 1999.
    [Conference Presentation, invited]
  5. R. Harasawa, J. Shikata, J. Suzuki, H. Imai, “Comparing the MOV and FR Reductions in Elliptic Curve Cryptography”, Lecture Note on Computer Science 1592, Advances in Cryptology Eurocrypt’99, pages 189-204, Springer-Verlag, May 1999.
    [Conference Paper]
  6. J. Suzuki.  “Learning Bayesian belief networks based on the MDL principle: An efficient algorithm using the branch and bound technique“. IEICE Trans. on Information and Systems E82-D(2): 356-367, Feb. 1999.
    [Journal paper]

1998

  1. J. Suzuki. “A relationship between context tree weighting and general model weighting techniques for tree sources“. IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E81-A(11): 2412-2417, November 1998.
    [Journal Paper]
  2. Kouki Hojo, Boris Ryabko, and Joe Suzuki, “It is not enough to assume stationary ergodic sources for analyzing universal coding”, International Symposium on Information Theory and its Applications, pages 113-116, Mexico City, Mexico, October 1998.
    [Conference Paper]
  3. Joseph H. Silverman and Joe Suzuki, “Elliptic Curve Discrete Logarithms and the Index Calculus”, Lecture Notes in Computer Science 1514, Advances in Cryptology-Asiacrypt, pages 110-125, October 1998.
    [Conference Paper]
  4. Joe Suzuki. “Universal Predicion and Universal Coding”, IMPRIM, Novosibirsk, Russia, June 1998.
    [Conference Presentation]
  5. J. Suzuki. “A further result on the Markov chain model of GAs and their application to SA-like strategy“. IEEE Trans. on Systems, Man, and Cybernetics, Part B (Cybernetics) 28(1): 95-192, January 1998.
    [Journal Paper]

1997

  1. Joe Suzuki, “Universal Codingand Universal Predicion”, IEEE International Symposium on Information Theory, Ulm, Germany,  June 1997.
    [Conference Presentation]
  2. Joe Suzuki, “On the Error Probability of Model Selection for Classification”, IEEE International Symposium on Information Theory, Ulm, Germany, June 1997.
    [Conference Presentation]
  3. Joe Suzuki,“On the Error Probability of Model Selection for Classification”, Artificial Intelligence and Statistics, pages 513-520, Fort Lauderdale, FL, January 1997.
    [Conference Paper]

1996

  1. Joe Suzuki,“Further Results on the Markov Chain Model of GAs and Their Application to SA-like Strategy”, Foundation of Genetic Algorithms, pages 53-72, San Diego, CA, August 1996
    [Conference Paper]
  2. Joe Suzuki, “Learning Bayesian Belief Networks Based on the Minimum Description Length Principle: an Efficient Algorithm using B &B Technique”, International Conference on Machine Learning, Bari, Italy, July 1996.
    [Conference Paper]
  3. Joe Suzuki, “A CTW Scheme for Non-tree Sources”, IEEE Data Compression Conference, Snowbird, Utah, April 1996.
    [Conference Presentation]

1995

  1. J. Suzuki. “Some notes on universal noiseless coding”. IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E78-A(12): 1840-1847, Dec. 1995.
    [Journal paper]
  2. Joe Suzuki, “A CTW Scheme for Some FSM Models”, IEEE International Symposium on Information Theory, Whistler, Canada. September 1995.
    [Conference Presentation]
  3. Joe Suzuki, “An Extension ‘An Extension on Learning Bayesian Belief Networks Based on MDL Principle’, IEEE International Symposium on Information Theory, Whistler, Canada, September 1995.
    [Conference Presentation]
  4. J. Suzuki. “A Markov chain analysis on simple genetic algorithms“. IEEE Trans. on Systems, Man, and Cybernetics 25(4): 655-659, April 1995.
    [Journal paper]

1994

  1. Joe Suzuki, “On a Generalized Context Tree Weighting Scheme”, The Fourth Benelux-Japan Workshop of Information Theory, Eindhoven, Neitherland, page 11, June 1994.[Conference Presentation]
  2. Joe Suzuki, “Tighter Bounds on Universal Noiseless Coding for Finite Sequences”, IEEE International Symposium on Information Theory,  Tronheim, Norway, page 388, June 1994.[Conference Presentation]
  3. Joe Suzuki, “A PAC Learning Theoretical Analysis on Software Testing”, IEEE International Workshop on Information Theory, Moscow, Russia, pages 98-101. July 1994
    [Conference Paper]
  4. Joe Suzuki, “Stochastic On-Line Prediction and MDL Principle”, Workshop on Applications of Descriptional Complexity to Inductive, Statistical, and Visual Inference, New Brunswick, NJ, July 1994.
    [Conference Presentation]

1993

  1. J. Suzuki. “A universal coding scheme based on minimizing minimax redundancy for sources with unknown model”.  IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E76-A(7): 1234-1239, July 1993.
    [Journal Paper]
  2. J. Suzuki. “Evaluations for estimation of information source based on state decomposition”, IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science, E76-A(7): 1240-1251, July 1993.
    [Journal Paper]
  3. Joe Suzuki, “A Markov Chain Analysis on Simple Genetic Algorithms”, International Conference on Genetic Algorithms, Arabama-Chanpaign, Illinois, July 1993.
    [Conference Paper]
  4. Joe Suzuki, “A Construction of Bayesian Networks from Databases on an MDL Principle”, The Ninth Conference on Uncertainty in Artificial Intelligence, Washington D. C., pages 266-273, July 1993.
    [Conference Paper]
  5. Joe Suzuki, “Minimizing Minimax Redundancy for Sources with Unknown Model”, IEEE International Symposium on Information Theory, San Antonio, Texas, January 1993.
    [Conference Presentation]

1992

  1. [Journal Paper (in Japanese)]
  2. [Journal Paper (in Japanese)]
  3. Joe Suzuki, “A Universal Coding Scheme Based on Minimizing Minimax Redundancy for Sources with Unknown Model”, International Symposium on Information Theory and its Applications, Singapore, November 1992
    [Conference Paper]
  4. Joe Suzuki, “Evaluations for Estimation of Information Source Based on State Decomposition”, CEMIT (International Conference on Economics / Managemant and Information Technology, Tokyo, Japan, August 1992
    [Conference Paper]

1990

  1. Joe Suzuki, “Generalization of the Learning Method for Classifying Rules with Consistency Irrespective of the Classified Patterns and the Representation Form”, International Symposium on Information Theory and its Applications, Waikiki, Hawaii, pages 495-498, November 1990.
    [Conference Paper]
  2. T. Matsushima, J. Suzuki, Inazumi H., and S. Hirasawa. “Inductive inference scheme at a nite stage of process from a viewpoint of source coding”. Transactions of IEICE E73(5): 644-652, May 1990.
    [Journal Paper]
  3. Toshiyasu Matsushima, Joe Suzuki, Hiroshige Inadumi, Shigeichi Hirasawa, “On the Optimal Inductive Inference Scheme from the Viewpoint of Source Coding”, IEEE International Symposium on Information Theory, San Diego, CA, January 1990.
    [Conference Presentation]

1989

  1. [Journal Paper (in Japanese)]

1988

  1. [Journal Paper (in Japanese)]
  2. Joe Suzuki, Toshiyasu Matsushima, Hiroshige Inadumi, Shigeich Hirasawa, “Feature Ordering and Stopping Rule Based on Maximizing Mutual Information”, IEEE International Symposium on Information Theory, Kobe, Japan, page 12, June, 1988.
    [Conference Presentation]
  3. Toshiyasu Matsushima, Joe Suzuki, Hiroshige Inazumi, Shigeichi Hirasawa, “On Uncertain Logic Based Upon Information Theory”, IEEE International Symposium on Information Theory, Kobe, Japan, page 133, June 1988.
    [Conference Presentation]

 

コメントを残す

メールアドレスが公開されることはありません。 * が付いている欄は必須項目です