In several agent-oriented scenarios in the real world, an autonomous agent that is situated in an unknown environment must learn through a process of trial and error to take actions that result in long-term benefit. Reinforcement Learning (or sequential decision making) is a paradigm well-suited to this requirement. Value function-based methods and policy search methods are contrasting approaches to solve reinforcement learning tasks. While both classes of methods benefit from independent theoretical analyses, these often fail to extend to the practical situations in which the methods are deployed. We conduct an emperical study to examine the strengths and weaknesses of these approaches by introducing a suite of test domains that can be varied for problem size, stochasticity, function approximation, and partial observability. Our results indicate clear patterns in the domain characteristics for which each class of methods excels. We investigate whether their strengths can be combine, and develop an approach to achieve that purpose. The effectiveness of this approach is also demonstrated on the challenging benchmark task of robot soccer Keepaway. We highlight several lines of inquiry that emanate from this study.
In The Eighth International Conference on Autonomous Agents and Multiagent Systems (AAMAS), pp. 749-756, Richland, SC, May 2009. International Foundation for Autonomous Agents and Multiagent Systems.

Shivaram Kalyanakrishnan Ph.D. Alumni shivaram [at] cs utexas edu
Peter Stone Faculty pstone [at] cs utexas edu