Jun-ichi TAKEUCHI

Polynomial Learnability of Stochastic Rules with Respect to the KL-Divergence and Quadratic Distance
Naoki ABE Jun-ichi TAKEUCHI Manfred K. WARMUTH 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 2001/03/01
Vol. E84-D  No. 3  pp. 299-316
Type of Manuscript:  PAPER
Category: Theory of Automata, Formal Language Theory
Keyword: 
PAC-learning, KL-divergence, quadratic-distance, stochastic rules, p-concepts, 
 Summary | Full Text:PDF(561KB)
Improved Sample Complexity Bounds for Parameter Estimation
Jun-ichi TAKEUCHI 
Publication:   IEICE TRANSACTIONS on Information and Systems
Publication Date: 1995/05/25
Vol. E78-D  No. 5  pp. 526-531
Type of Manuscript:  Special Section PAPER (Special Issue on Algorithmic Learning Theory)
Category: Computational Learning Theory
Keyword: 
PAC learning, sample complexity, Laplace estimator, Kullback-Leibler divergence, 
 Summary | Full Text:PDF(415.8KB)