Automatic generation of macro-actions using genetic algorithm for reinforcement learning

被引：0

作者：

Tateyama, T ^{[1
]}

Kawata, S ^{[1
]}

Oguchi, T ^{[1
]}

机构：

[1] Tokyo Metropolitan Univ, Grad Sch Engn, Hachioji, Tokyo 1920397, Japan

来源：

SICE 2002: PROCEEDINGS OF THE 41ST SICE ANNUAL CONFERENCE, VOLS 1-5 | 2002年

关键词：

reinforcement learning; macro-action; Semi-Markov Decision Processes; classifier system;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The main problem of reinforcement learning is that learning converges slowly. As one of the solution, McGovern proposed "macro-action". However, a human expert needs to design macro-actions which adapt to an environment. In this paper, we propose a new method that enables to generate the macro-actions which adapt to the enviroment automatically using genetic algorithm.

引用

页码：286 / 289

页数：4

共 50 条

[1] Automatic construction and evaluation of macro-actions in reinforcement learning
Farahani, Marzieh Davoodabadi
Mozayani, Nasser
[J]. APPLIED SOFT COMPUTING, 2019, 82
[2] Learning macro-actions in reinforcement learning
Randlov, J
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 11, 1999, 11 : 1045 - 1051
[3] An acquiring method of macro-actions in reinforcement learning
Yoshikawa, Takeshi
Kurihara, Masahito
[J]. 2006 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-6, PROCEEDINGS, 2006, : 4813 - +
[4] A Method for Learning Macro-Actions for Virtual Characters Using Programming by Demonstration and Reinforcement Learning
Sung, Yunsick
Cho, Kyungeun
[J]. JOURNAL OF INFORMATION PROCESSING SYSTEMS, 2012, 8 (03): : 409 - 420
[5] Multi-Agent/Robot Deep Reinforcement Learning with Macro-Actions
Xiao, Yuchen
Hoffman, Joshua
Xia, Tian
Amato, Christopher
[J]. THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2020, 34 : 13965 - 13966
[6] A memory-based reinforcement learning model utilizing macro-actions
Murata, M
Ozawa, S
[J]. ADAPTIVE AND NATURAL COMPUTING ALGORITHMS, 2005, : 78 - 81
[7] A Reinforcement Learning Model Using Macro-actions in Multi-task Grid-World Problems
Onda, Hiroshi
Ozawa, Seiichi
[J]. 2009 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC 2009), VOLS 1-9, 2009, : 3088 - 3093
[8] Strategic Attentive Writer for Learning Macro-Actions
Vezhnevets, Alexander
Mnih, Volodymyr
Agapiou, John
Osindero, Simon
Graves, Alex
Vinyals, Oriol
Kavukcuoglu, Koray
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 29 (NIPS 2016), 2016, 29
[9] Implicit Learning of Compiled Macro-Actions for Planning
Newton, M. A. Hakim
Levine, John
[J]. ECAI 2010 - 19TH EUROPEAN CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2010, 215 : 323 - 328
[10] MAGIC: Learning Macro-Actions for Online POMDP Planning
Lee, Yiyuan
Cai, Panpan
Hsu, David
[J]. ROBOTICS: SCIENCE AND SYSTEM XVII, 2021,

← 1 2 3 4 5 →