Presentation | 2004/10/12 Reinforcement Learning with State Segmentation Method based on Sensory Change Anticipations Hisashi HANDA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The situatedness is one of the most important notion for constructing state segmentation of the reinforcement learning agents. Hence, we propose a new state segmentation method referring to sensation-action series. The proposed method quantizes input space, and anticipates the next inputs as a consequence of actions of agents. Moreover, the proposed method constitutes the state segmentation of agents based on anticipation accuracy. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | reinforcement learning / growing neural gas / incremental state segmentation method |
Paper # | NC2004-81 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2004/10/12(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) | Reinforcement Learning with State Segmentation Method based on Sensory Change Anticipations |
Sub Title (in English) | |
Keyword(1) | reinforcement learning |
Keyword(2) | growing neural gas |
Keyword(3) | incremental state segmentation method |
1st Author's Name | Hisashi HANDA |
1st Author's Affiliation | Faculty of Engineering, Okayama University() |
Date | 2004/10/12 |
Paper # | NC2004-81 |
Volume (vol) | vol.104 |
Number (no) | 349 |
Page | pp.pp.- |
#Pages | 5 |
Date of Issue |