Art

J-GLOBAL ID：201202254172432238   Reference number：12A1064302

On a Non-asymptotic Analysis Using Large Deviation Principles in the Multiarmed Bandit Problem

多腕バンディット問題における大偏差原理を用いた非漸近的解析について

Author (2)： HONDA Junya (Univ. Tokyo, Chiba, JPN) ,  TAKEMURA Akimichi (Univ. Tokyo, Tokyo, JPN)
Material： 電子情報通信学会技術研究報告  (IEICE Technical Report (Institute of Electronics, Information and Communication Engineers))
Volume： 112  Issue： 83(IBISML2012 1-12)  Page： 65-72  Publication year： Jun. 12, 2012 
JST Material Number： S0532B  ISSN： 0913-5685  Document type： Proceedings
Article type： 短報  Country of issue： Japan (JPN)  Language： ENGLISH (EN)

Thesaurus term： *reinforcement learning ,  modeling ,  formulation ,  performance evaluation ,  wage ,  human behavior ,  probability distribution ,  entropy ,  information content
Semi thesaurus term： Kullback-Leibler divergence ,  gamble ,  dilemma ,  modeling ,  Reward

JST classification (1)： Artificial intelligence  (JE08000Z)

Reference (17)：

• GITTINS, J. C. Multi-armed bandit allocation indices. 1989
• AGRAWAL, R. The continuum-armed bandit problem. SIAM Journal on Control and Optimization. 1995, 33, 6, 1926-1951
• AUER, P. The nonstochastic multiarmed bandit problem. SIAM Journal on Computing. 2002, 32, 1, 48-77
• ROBBINS, H. Some aspects of the sequential design of experiments. Bulletin of the American Mathematical Society. 1952, 58, 5, 527-535
• LAI, T. L. Asymptotically efficient adaptive allocation rules. Advances in Applied Mathematics. 1985, 6, 4-22

Terms in the title (3)： 腕 ,  大偏差原理 ,  解析