Provably Efficient Multi-Task Reinforcement Learning with Model Transfer

Abstract

We study multi-task reinforcement learning (RL) in tabular episodic Markov decision processes (MDPs). We formulate a heterogeneous multi-player RL problem, in which a group of players concurrently face similar but not necessarily identical MDPs, with a goal of improving their collective performance through inter-player information sharing. We design and analyze a model-based algorithm, and provide gap-dependent and gap-independent regret upper and lower bounds that characterize the intrinsic complexity of the problem.

PDF NeurIPS OpenReview Semantic Scholar

Cite

Text

Zhang and Wang. "Provably Efficient Multi-Task Reinforcement Learning with Model Transfer." Neural Information Processing Systems, 2021.

Markdown

[Zhang and Wang. "Provably Efficient Multi-Task Reinforcement Learning with Model Transfer." Neural Information Processing Systems, 2021.](https://mlanthology.org/neurips/2021/zhang2021neurips-provably/)

BibTeX

@inproceedings{zhang2021neurips-provably,
  title     = {{Provably Efficient Multi-Task Reinforcement Learning with Model Transfer}},
  author    = {Zhang, Chicheng and Wang, Zhi},
  booktitle = {Neural Information Processing Systems},
  year      = {2021},
  url       = {https://mlanthology.org/neurips/2021/zhang2021neurips-provably/}
}