Constructing continuous action space from basis functions for fast and stable reinforcement learning

Akihiko Yamaguchi, Jun Takamatsu, Tsukasa Ogasawara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper presents a new continuous action space for reinforcement learning (RL) with the wire-fitting [1]. The wire-fitting has a desirable feature to be used with action value function based RL algorithms. However, the wire-fitting becomes unstable caused by changing the parameters of actions. Furthermore, the acquired behavior highly depend on the initial values of the parameters. The proposed action space is expanded from the DCOB, proposed by Yamaguchi et al. [2], where the discrete action set is generated from given basis functions. Based on the DCOB, we apply some constraints to the parameters in order to obtain stability. Furthermore, we also describe a proper way to initialize the parameters. The simulation results demonstrate that the proposed method outperforms the wire-fitting. On the other hand, the resulting performance of the proposed method is the same as, or inferior to the DCOB. This paper also discuss about this result.

Original languageEnglish
Title of host publicationRO-MAN 2009 - 18th IEEE International Symposium on Robot and Human Interactive
Pages401-407
Number of pages7
DOIs
Publication statusPublished - 2009 Dec 1
Externally publishedYes
Event18th IEEE International Symposium on Robot and Human Interactive, RO-MAN 2009 - Toyama, Japan
Duration: 2009 Sep 272009 Oct 2

Publication series

NameProceedings - IEEE International Workshop on Robot and Human Interactive Communication

Other

Other18th IEEE International Symposium on Robot and Human Interactive, RO-MAN 2009
CountryJapan
CityToyama
Period09/9/2709/10/2

Keywords

  • Continuous action space
  • Crawling
  • Jumping
  • Motion learning
  • Reinforcement learning

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Human-Computer Interaction

Fingerprint Dive into the research topics of 'Constructing continuous action space from basis functions for fast and stable reinforcement learning'. Together they form a unique fingerprint.

Cite this