Art

J-GLOBAL ID：201202274916773803   Reference number：12A0422030

Analysis of Time Series Data Accompanied with Rewards and Actions using Reinforcement Learning

報酬と行動決定を伴う時系列データの強化学習を用いたオフライン分析

Author (5)： ASO HIDEKI (産業技術総合研) ,  SHIRO MASANORI (産業技術総合研) ,  KAMISHIMA TOSHIHIRO (産業技術総合研) ,  AKAHO SHOTARO (産業技術総合研) ,  KORO TAKAHIDE (東大 医 病院)
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： 111  Issue： 419(NC2011 97-120)  Page： 107-112  Publication year： Jan. 19, 2012 
JST Material Number： S0532B  ISSN： 0913-5685  Document type： Proceedings
Article type： 短報  Country of issue： Japan (JPN)  Language： JAPANESE (JA)

Thesaurus term： data ,  *time series analysis ,  *reinforcement learning ,  Markov process ,  data processing ,  *data mining ,  agent ,  evaluation function ,  cardiovascular disease ,  medical record ,  possibility ,  experiment
Semi thesaurus term： MDP ,  offline processing ,  time series data ,  intelligent agent ,  possibility of application ,  partially observable Markov decision process ,  evaluation experiment

JST classification (1)： Artificial intelligence  (JE08000Z)

Reference (15)：

• ENGEL, Y. Bayes meets Bellman : the Gaussian process approach to temporal difference learning. Proceedings of ICML-2003. 2003
• HAUSKRECHT, M. Planning treatment of ischemic heart disease with partially observable Markov decision processes. Artificial Intelligence in Medicine. 2000, 18, 221-244
• LEVIN, E. Using Markov decision processes for learning dialogue strategies. IEEE Transactions on Speech and Audio Processing. 1998, 8, 11-23
• NG, A. Y. Algorithms for inverse reinforcement learning. Proceedings of 17th International Conference on Machine Learning, 2000. 2000, 663-670
• PAEK, T. Reinforcement learning for spoken dialogue systems : comparing strength and weaknesses for practical deployment. 2006

Terms in the title (6)： 報酬 ,  行動決定 ,  時系列データ ,  強化学習 ,  オフライン ,  分析