Skip to main content Skip to main navigation


Efficient data reuse in value function approximation

Hirotaka Hachiya; Takayuki Akiyama; Masashi Sugiyama; Jan Peters
In: 2009 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning (ADPRL 2009) Proceedings. IEEE Symposium on Adaptive Dynamic Programming And Reinforcement Learning (ADPRL-2009), March 30 - April 2, Nashville, TN, USA, Pages 8-15, IEEE Symposium Series on Computational Intelligence, ISBN 978-1-4244-2761-1, IEEE, 2009.


Off-policy reinforcement learning is aimed at efficiently using data samples gathered from a policy that is different from the currently optimized policy. A common approach is to use importance sampling techniques for compensating for the bias of value function estimators caused by the difference between the data-sampling policy and the target policy. However, existing off-policy methods often do not take the variance of the value function estimators explicitly into account and therefore their performance tends to be unstable. To cope with this problem, we propose using an adaptive importance sampling technique which allows us to actively control the trade-off between bias and variance. We further provide a method for optimally determining the trade-off parameter based on a variant of cross-validation. The usefulness of the proposed approach is demonstrated through simulated swing-up inverted-pendulum problem.

Weitere Links