AIPS
2011
13 years 4 months ago
2011
In this paper, we present a new algorithm that integrates recent advances in solving continuous bandit problems with sample-based rollout methods for planning in Markov Decision P...