Policy Learning Using Modified Learning Vector Quantization for Reinforcement Learning Problems

DOI HANDLE オープンアクセス
  • Afif Mohd Faudzi Ahmad
    Department of Electrical and Electronic Engineering, Graduate School of Information and Electrical Engineering, Kyushu University | Department of Electrical and Electronic Engineering, Universiti Malaysia
  • 村田 純一
    九州大学大学院システム情報科学研究院電気システム工学 : 教授

この論文をさがす

抄録

Reinforcement learning (RL) enables an agent to _nd an optimal solution to a problem by interacting with the environment. In the previous research, Q-learning, one of the popular learning meth-ods in RL, is used to generate a policy. From it, abstract policy is extracted by LVQ algorithm. In this paper, the aim is to train the agent to learn an optimal policy from scratch as well as to generate the abstract policy in a single operation by LVQ algorithm. When applying LVQ algorithm in a RL frame-work, due to an erroneous teaching signal in LVQ algorithm, the learning sometimes end up with failure or with non-optimal solution. Here, a new LVQ algorithm is proposed to overcome this problem. The new LVQ algorithm introduce, _rst, a regular reward that is obtained by the agent autonomously based on its behavior and second, a function that convert a regular reward to a new reward so that the learning system does not su_er from an undesirable e_ect by a small reward. Through these modi_cations, the agent is expected to _nd the optimal solution more e_ciently.

収録刊行物

キーワード

詳細情報 詳細情報について

  • CRID
    1390572174796264192
  • NII論文ID
    120005697190
  • NII書誌ID
    AN10569524
  • DOI
    10.15017/1560523
  • ISSN
    21880891
    13423819
  • HANDLE
    2324/1560523
  • 本文言語コード
    en
  • データソース種別
    • JaLC
    • IRDB
    • CiNii Articles
  • 抄録ライセンスフラグ
    使用可

問題の指摘

ページトップへ