The ambitious goal of transfer learning is to accelerate learning on a target task after training on a different, but related, source task. While many past transfer methods have focused on transferring value-functions, this paper presents a method for transferring policies across tasks with different state and action spaces. In particular, this paper utilizes transfer via inter-task mappings for policy search methods (sc tvitm-ps) to construct a transfer functional that translates a population of neural network policies trained via policy search from a source task to a target task. Empirical results in robot soccer Keepaway and Server Job Scheduling show that sc tvitm-ps can markedly reduce learning time when full inter-task mappings are available. The results also demonstrate that sc tvitm-ps still succeeds when given only incomplete inter-task mappings. Furthermore, we present a novel method for emphlearning such mappings when they are not available, and give results showing they perform comparably to hand-coded mappings.

PDF, PS, HTML

In Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems, May 2007.

@InProceedings{taylor:ijcaams07, title={Transfer via Inter-Task Mappings in Policy Search Reinforcement Learning}, author={Matthew E. Taylor and Shimon Whiteson and Peter Stone}, booktitle={Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems}, month={May}, url="http://www.cs.utexas.edu/users/ai-lab?taylor:ijcaams07", year={2007} }