| Presentation | 2007-01-25 Adaptation of a Reinforcement Learning System IP MBN Using a Clustering Algorithm to Environmental Changes Daisuke KITAKOSHI, Terumasa YAMAGUCHI, Hiroyuki SHIOYA, Ryohei NAKANO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We have proposed an adaptive learning system IPMBN using a mixture of Bayesian networks for agents, and also have investigated its adjustability to dynamic environments. Each component of the mixture (i.e. a Bayesian network) can be regarded as a stochastic knowledge representation corresponding to a policy in environment. The performance to represent the environmental information is considered to increase with the number of BNs incorporated into the mixture; however, greater computational cost is required to constitute the mixture. It is thus disirable that our system can render the environmental information with as less BNs as possible. In this paper, we introduce a clustering algorithm into IPMBN to extract BNs, which are applicable for representing a variety of policies, out of a quantity of BNs, and then discuss the performance of the system. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Mixture of Bayesian Networks / Profit Sharing / Hellinger Distance / Clustering |
| Paper # | NC2006-99 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2007/1/18(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Neurocomputing (NC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Adaptation of a Reinforcement Learning System IP MBN Using a Clustering Algorithm to Environmental Changes |
| Sub Title (in English) | |
| Keyword(1) | Mixture of Bayesian Networks |
| Keyword(2) | Profit Sharing |
| Keyword(3) | Hellinger Distance |
| Keyword(4) | Clustering |
| 1st Author's Name | Daisuke KITAKOSHI |
| 1st Author's Affiliation | Graduate School of Engineering, Nagoya Institute of Technology() |
| 2nd Author's Name | Terumasa YAMAGUCHI |
| 2nd Author's Affiliation | Faculty of Engineering, Muroran Institute of Technology |
| 3rd Author's Name | Hiroyuki SHIOYA |
| 3rd Author's Affiliation | Faculty of Engineering, Muroran Institute of Technology |
| 4th Author's Name | Ryohei NAKANO |
| 4th Author's Affiliation | Graduate School of Engineering, Nagoya Institute of Technology |
| Date | 2007-01-25 |
| Paper # | NC2006-99 |
| Volume (vol) | vol.106 |
| Number (no) | 500 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |