Incorporating Domain Models into Bayesian Optimization for Reinforcement Learning

Abstract

In many Reinforcement Learning (RL) domains there is a high cost for generating experience in order to evaluate an agent’s performance. An appealing approach to reducing the number of expensive evaluations is Bayesian Optimization (BO), which is a framework for global optimization of noisy and costly to evaluate functions. Prior work in a number of RL… (More)

Topics

4 Figures and Tables

Cite this paper

@inproceedings{Wilson2010IncorporatingDM, title={Incorporating Domain Models into Bayesian Optimization for Reinforcement Learning}, author={Aaron Wilson and Alan Fern and Prasad Tadepalli}, year={2010} }