@article{conservativeqimprovementreinforcement,
  title = {Conservative Q-Improvement: Reinforcement Learning for an Interpretable Decision-Tree Policy},
  author = {Aaron M. Roth and Nicholay Topin and Pooyan Jamshidi and Manuela Veloso},
  year = {2019},
  eprint = {1907.01180},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1907.01180v1},
}