Representation Transfer for Reinforcement Learning
Matthew Edmund Taylor, Peter Stone · 2007
Abstract Transfer learning problems are typically framed as leveragingknowledge learned on a source task to improve learning on a related, but different, target task. Current transfer learningmethods are able to successfully transfer knowledge from a source reinforcement learning task into a target task, reducinglearning time. However, the complimentary task of transferring knowledge between agents with different internal repre-sentations has not been well explored The goal in both types of transfer problems is the same: reduce the time needed tolearn the target with transfer, relative to learning the target without transfer. This work defines representation transfer,contrasts it with task transfer, and introduces two novel algorithms. Additionally, we show representation transfer al-gorithms can also be successfully used for task transfer, providing an empirical connection between the two problems.These algorithms are fully implemented in a complex multiagent domain and experiments demonstrate that transferringthe learned knowledge between different representations is both possible and beneficial.