Rollout Sampling Approximate Policy Iteration

Details Rollout Sampling Approximate Policy Iteration Rollout Sampling Approximate Policy Iteration

2008-05-14

arxiv.org/abs/0805.2027v2

Computer Science

Learning

18 pages, 2 figures, to appear in Machine Learning 72(3). Presented at EWRL08, to be presented at ECML 2008

Scientific paper

10.1007/s10994-008-5069-3

Several researchers have recently investigated the connection between reinforcement learning and classification. We are motivated by proposals of approximate policy iteration schemes without value functions which focus on policy representation using classifiers and address policy learning as a supervised learning problem. This paper proposes variants of an improved policy iteration scheme which addresses the core sampling problem in evaluating a policy through simulation as a multi-armed bandit machine. The resulting algorithm offers comparable performance to the previous algorithm achieved, however, with significantly less computational effort. An order of magnitude improvement is demonstrated experimentally in two standard reinforcement learning domains: inverted pendulum and mountain-car.

Affiliated with

Also associated with

No associations

Rating

Rollout Sampling Approximate Policy Iteration does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Rollout Sampling Approximate Policy Iteration, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Rollout Sampling Approximate Policy Iteration will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-55700