• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
Shimon Whiteson, Brian Tanner, Matthew
E. Taylor, and Peter Stone. Generalized Domains for Empirical Evaluations
in Reinforcement Learning. In ICML Workshop on Evaluation Methods for Machine Learning, June 2009.
Workshop
page
Many empirical results in reinforcement learning are based on a very small set of environments. These results often represent the best algorithm parameters that were found after an ad-hoc tuning or fitting process. We argue that presenting tuned scores from a small set of environments leads to method overfitting, wherein results may not generalize to similar environments. To address this problem, we advocate empirical evaluations using generalized domains: parameterized problem generators that explicitly encode variations in the environment to which the learner should be robust. We argue that evaluating across a set of these generated problems offers a more meaningful evaluation of reinforcement learning algorithms.
@InProceedings(ICML09ws-shimon, author="Shimon Whiteson and Brian Tanner and Matthew E.\ Taylor and Peter Stone", title="Generalized Domains for Empirical Evaluations in Reinforcement Learning", booktitle="{ICML} Workshop on Evaluation Methods for Machine Learning", month="June", year="2009", abstract={ Many empirical results in reinforcement learning are based on a very small set of environments. These results often represent the best algorithm parameters that were found after an ad-hoc tuning or fitting process. We argue that presenting tuned scores from a small set of environments leads to method overfitting, wherein results may not generalize to similar environments. To address this problem, we advocate empirical evaluations using generalized domains: parameterized problem generators that explicitly encode variations in the environment to which the learner should be robust. We argue that evaluating across a set of these generated problems offers a more meaningful evaluation of reinforcement learning algorithms. }, wwwnote={<a href="http://www.site.uottawa.ca/ICML09WS/">Workshop page</a>}, )
Generated by bib2html.pl (written by Patrick Riley ) on Fri May 13, 2011 08:07:35