Genetic Reinforcement Learning for Neurocontrol Problems

  • Authors:

  • Darrell Whitley;Stephen Dominic;Rajarshi Das;Charles W. Anderson

  • Affiliations:

  • Computer Science Department, Colorado State University, Fort Collins, CO 80523. whitley@cs.colostate.edu;Computer Science Department, Colorado State University, Fort Collins, CO 80523;Computer Science Department, Colorado State University, Fort Collins, CO 80523;Computer Science Department, Colorado State University, Fort Collins, CO 80523

  • Venue:
  • Machine Learning - Special issue on genetic algorithms
  • Year:
  • 1993

Quantified Score

Hi-index 0.01

Visualization

Abstract

Empirical tests indicate that at least one class of genetic algorithms yields good performance for neural network weight optimization in terms of learning rates and scalability. The successful application of these genetic algorithms to supervised learning problems sets the stage for the use of genetic algorithms in reinforcement learning problems. On a simulated inverted-pendulum control problem, “genetic reinforcement learning” produces competitive results with AHC, another well-known reinforcement learning paradigm for neural networks that employs the temporal difference method. These algorithms are compared in terms of learning rates, performance-based generalization, and control behavior over time.