Piecewise Linear Parametrization of Policies: Towards Interpretable Deep Reinforcement Learning.

Maxime Wabartha, Joelle Pineau

Browse the full ICLR paper archive.