Shimon Whiteson's Publications

Sorted by DateClassified by Publication Type

Evolutionary Function Approximation for Reinforcement Learning

Shimon Whiteson. Evolutionary Function Approximation for Reinforcement Learning. In GECCO 2006: Proceedings of the Genetic and Evolutionary Computation Conference Graduate Student Workshop, July 2006. Best Paper Award, Graduate Student Workshop.

Download

[PDF]1.0MB  

Abstract

Temporal difference methods are theoretically grounded and empirically effective methods for addressing reinforcement learning problems. In most real-world reinforcement learning tasks, TD methods require a function approximator to represent the value function. However, using function approximators requires manually making crucial representational decisions. This thesis investigates evolutionary function approximation, a novel approach to automatically selecting function approximator representations that enable efficient individual learning. This method evolves individuals that are better able to learn. I present a fully implemented instantiation of evolutionary function approximation which combines NEAT, a neuroevolutionary optimization technique, with Q-learning, a popular TD method. The resulting NEAT+Q algorithm automatically discovers effective representations for neural network function approximators. This thesis also presents on-line evolutionary computation, which improves the on-line performance of evolutionary computation by borrowing selection mechanisms used in TD methods to choose individual actions and using them in evolutionary computation to select policies for evaluation. I evaluate these contributions with extended empirical studies in two domains: 1) the mountain car task, a standard reinforcement learning benchmark on which neural network function approximators have previously performed poorly and 2) server job scheduling, a large probabilistic domain drawn from the field of autonomic computing. The results demonstrate that evolutionary function approximation can significantly improve the performance of TD methods and on-line evolutionary computation can significantly improve evolutionary methods.

BibTeX Entry

@InProceedings{whiteson:gecco06ws,
  author       = "Shimon Whiteson",
  title	       = "Evolutionary Function Approximation for Reinforcement Learning",
  booktitle    = "GECCO 2006: Proceedings of the Genetic and
                  Evolutionary Computation Conference Graduate Student Workshop",
  month	       = "July",
  year	       = 2006,
  note         = "\textbf{Best Paper Award, Graduate Student Workshop}.",
}

Generated by bib2html.pl (written by Patrick Riley ) on Wed Mar 07, 2012 15:56:00