Reinforcement learning using on-line EM algorithm

Bibliographic Information

Other Title
  • オンラインEMアルゴリズムを用いた強化学習法

Search this article

Description

In this research report, we propose a new reinforcement learning (RL) method based on an actor-critic architecture. The actor and the critic are approximated by normalized Gausssian networks, which are trained by the on-line EM algorithm proposed in our previous paper. We apply our RL method to the task of swing-up and stabilizing a single pendulum and the task of balacing a double pendulum near the upright position. The experimental results show that our RL method can be applied to optimal control problems having continuous state/action spaces.

Journal

Citations (2)*help

See more

References(13)*help

See more

Details 詳細情報について

  • CRID
    1570291227540510080
  • NII Article ID
    110003233550
  • NII Book ID
    AN10091178
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top