Automatically Mapped Transfer Between Reinforcement Learning Tasks via Three-Way Restricted Boltzmann Machines

H. B. Ammar, D. C. Mocanu, M. E. Taylor, K. Driessens, K. Tuyls, G. Weiss

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

8 Downloads (Pure)


Existing reinforcement learning approaches are often hampered by learning tabula rasa. Transfer for reinforcement learning tackles this problem by enabling the reuse of previously learned results, but may require an inter-task mapping to encode how the previously learned task and the new task are related. This paper presents an autonomous framework for learning inter-task mappings based on an adaptation of restricted boltzmann machines. Both a full model and a computationally efficient factored model are introduced and shown to be effective in multiple transfer learning scenarios.keywordstransfer learningreinforcement learninginter-task mappingboltzmann machinesleast squares policy iteration.
Original languageEnglish
Title of host publicationProceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD 2013)
Number of pages16
Publication statusPublished - 2013

Publication series

SeriesLecture Notes in Computer Science

Cite this