PAC Reinforcement Learning without Real-World Feedback

Yuren Zhong; Aniket Anand Deshmukh; Clayton Scott

PAC Reinforcement Learning without Real-World Feedback

Machine Learning 2019-10-28 v3 Machine Learning

Authors: Yuren Zhong , Aniket Anand Deshmukh , Clayton Scott

Abstract

This work studies reinforcement learning in the Sim-to-Real setting, in which an agent is first trained on a number of simulators before being deployed in the real world, with the aim of decreasing the real-world sample complexity requirement. Using a dynamic model known as a rich observation Markov decision process (ROMDP), we formulate a theoretical framework for Sim-to-Real in the situation where feedback in the real world is not available. We establish real-world sample complexity guarantees that are smaller than what is currently known for directly (i.e., without access to simulators) learning a ROMDP with feedback.

Keywords

sim-to-real transfer reinforcement learning multi-agent reinforcement learning

Cite

@article{arxiv.1909.10449,
  title  = {PAC Reinforcement Learning without Real-World Feedback},
  author = {Yuren Zhong and Aniket Anand Deshmukh and Clayton Scott},
  journal= {arXiv preprint arXiv:1909.10449},
  year   = {2019}
}

Related papers

View all related →