@inproceedings{amaximumentropyapproachtooffpolicy,
  title = {A maximum-entropy approach to off-policy evaluation in average-reward MDPs},
  author = {Nevena Lazic and Dong Yin and Mehrdad Farajtabar and Nir Levine and Dilan Gorur and Chris Harris and Dale Schuurmans},
  year = {2020},
  booktitle = {NeurIPS 2020 12},
  eprint = {2006.12620},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2006.12620v1},
}