@article{deterministicpolicyoptimizationby, title = {Deterministic Policy Optimization by Combining Pathwise and Score Function Estimators for Discrete Action Spaces}, author = {Daniel Levy and Stefano Ermon}, year = {2017}, eprint = {1711.08068}, archivePrefix = {arXiv}, url = {http://arxiv.org/abs/1711.08068v1}, }