Sumio WATANABE

Testing Homogeneity for Normal Mixture Models: Variational Bayes Approach
Natsuki KARIYA Sumio WATANABE 
Publication:   
Publication Date: 2020/11/01
Vol. E103-A  No. 11  pp. 1274-1282
Type of Manuscript:  PAPER
Category: Information Theory
Keyword: 
hypothesis test, Bayesian statistics, variational inference, singular model, mixture model, likelihood ratio, 
 Summary | Full Text:PDF(962.7KB)
Statistical Learning Theory of Quasi-Regular Cases
Koshi YAMADA Sumio WATANABE 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2012/12/01
Vol. E95-A  No. 12  pp. 2479-2487
Type of Manuscript:  PAPER
Category: General Fundamentals and Boundaries
Keyword: 
statistical learning theory, quasi-regular case, birational invariants, generalization error, 
 Summary | Full Text:PDF(339.1KB)
Equations of States in Statistical Learning for an Unrealizable and Regular Case
Sumio WATANABE 
Publication:   IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2010/03/01
Vol. E93-A  No. 3  pp. 617-626
Type of Manuscript:  PAPER
Category: Neural Networks and Bioengineering
Keyword: 
bayes learning, generalization, information criterion, singular, regular, 
 Summary | Full Text:PDF(160.3KB)
Generalization Performance of Subspace Bayes Approach in Linear Neural Networks
Shinichi NAKAJIMA Sumio WATANABE 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 2006/03/01
Vol. E89-D  No. 3  pp. 1128-1138
Type of Manuscript:  PAPER
Category: Algorithm Theory
Keyword: 
empirical Bayes, variational Bayes, neural networks, reduced-rank regression, James-Stein, unidentifiable, 
 Summary | Full Text:PDF(330.2KB)
A Modified Information Criterion for Automatic Model and Parameter Selection in Neural Network Learning
Sumio WATANABE 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 1995/04/25
Vol. E78-D  No. 4  pp. 490-499
Type of Manuscript:  PAPER
Category: Bio-Cybernetics and Neurocomputing
Keyword: 
information criterion, AIC, MDL, weight pruning, prediction error, generalized learning, 
 Summary | Full Text:PDF(765.5KB)