Efficient Learning and Planning Within the Dyna Framework

Peng Jing, Ronald J. Williams

Research output: Contribution to journalArticle

71 Scopus citations

Abstract

Sutton’s Dyna framework provides a novel and computationally appealing way to integrate learning, planning, and reacting in autonomous agents. Examined here is a class of strategies designed to enhance the learning and planning power of Dyna systems by increasing their computational efficiency. The benefit of using these strategies is demonstrated on some simple abstract learning tasks.

Original languageEnglish
Pages (from-to)437-454
Number of pages18
JournalAdaptive Behavior
Volume1
Issue number4
DOIs
StatePublished - Mar 1993

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Experimental and Cognitive Psychology
  • Behavioral Neuroscience

Keywords

  • dynamic programming
  • reinforcement learning
  • sequential decision problems

Cite this