Efficient Learning and Planning Within the Dyna Framework

1 March 1993

journal article
Published by SAGE Publications in Adaptive Behavior

Vol. 1 (4) , 437-454
https://doi.org/10.1177/105971239300100403

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.

Keywords

REACTING
SUTTON'S
EFFICIENT
AUTONOMOUS
APPEALING
COMPUTATIONALLY
DYNA FRAMEWORK
LEARNING AND PLANNING

This publication has 4 references indexed in Scilit:

Q-learning
Machine Learning, 1992
Reinforcement learning is direct adaptive optimal control
IEEE Control Systems, 1992
Learning to predict by the methods of temporal differences
Machine Learning, 1988
Some Studies in Machine Learning Using the Game of Checkers
IBM Journal of Research and Development, 1959