Online Learning of Genetic Network Programming and its Application to Prisoner's Dilemma Game.
-
- Mabu Shingo
- Department of Electrical and Electronic Systems Engineering,Kyushu University
-
- Hirasawa Kotaro
- Graduate School of Information, Production, and Systems,Waseda University
-
- Hu Jinglu
- Department of Electrical and Electronic Systems Engineering,Kyushu University
-
- Murata Junichi
- Department of Electrical and Electronic Systems Engineering,Kyushu University
この論文をさがす
抄録
A new evolutionary model with the network structure named Genetic Network Programming (GNP) has been proposed recently. GNP, that is, an expansion of GA and GP, represents solutions as a network structure and evolves it by using “offline learning (selection, mutation, crossover)”. GNP can memorize the past action sequences in the network flow, so it can deal with Partially Observable Markov Decision Process (POMDP) well. In this paper, in order to improve the ability of GNP, Q learning (an off-policy TD control algorithm) that is one of the famous online methods is introduced for online learning of GNP. Q learning is suitable for GNP because (1) in reinforcement learning, the rewards an agent will get in the future can be estimated, (2) TD control doesn’t need much memory and can learn quickly, and (3) off-policy is suitable in order to search for an optimal solution independently of the policy. Finally, in the simulations, online learning of GNP is applied to a player for “Prisoner’s dilemma game” and its ability for online adaptation is confirmed.
収録刊行物
-
- 電気学会論文誌C(電子・情報・システム部門誌)
-
電気学会論文誌C(電子・情報・システム部門誌) 123 (3), 535-543, 2003
一般社団法人 電気学会
- Tweet
キーワード
詳細情報 詳細情報について
-
- CRID
- 1390001204606305024
-
- NII論文ID
- 130000089347
- 30011601796
-
- NII書誌ID
- AN10065950
-
- ISSN
- 13488155
- 03854221
-
- NDL書誌ID
- 6480654
-
- 本文言語コード
- en
-
- データソース種別
-
- JaLC
- NDL
- Crossref
- CiNii Articles
-
- 抄録ライセンスフラグ
- 使用不可