Decision-Making with Non-Markovian Rewards: From LTL to automata-based reward shaping. Camacho, A., Chen, O., Sanner, S., & McIlraith, S. A. In Proceedings of the Multi-disciplinary Conference on Reinforcement Learning and Decision Making (RLDM-17), pages 279-283, 2017. See also University of Toronto Technical Report CSRG-632
Decision-Making with Non-Markovian Rewards: From LTL to automata-based reward shaping [pdf]Paper  bibtex   5 downloads  

Downloads: 5