@inproceedings{sampleefficientactorcriticreinforcement,
  title = {Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management},
  author = {Pei-Hao Su and Pawel Budzianowski and Stefan Ultes and Milica Gasic and Steve Young},
  year = {2017},
  booktitle = {WS 2017 8},
  eprint = {1707.00130},
  archivePrefix = {arXiv},
  url = {http://arxiv.org/abs/1707.00130v2},
}