Paper ID: 2207.06269

Policy Optimization with Sparse Global Contrastive Explanations

Jiayu Yao, Sonali Parbhoo, Weiwei Pan, Finale Doshi-Velez

We develop a Reinforcement Learning (RL) framework for improving an existing behavior policy via sparse, user-interpretable changes. Our goal is to make minimal changes while gaining as much benefit as possible. We define a minimal change as having a sparse, global contrastive explanation between the original and proposed policy. We improve the current policy with the constraint of keeping that global contrastive explanation short. We demonstrate our framework with a discrete MDP and a continuous 2D navigation domain.

Submitted: Jul 13, 2022