Batch Policy Learning under Constraints

Abstract

When learning policies for real-world domains, two important questions arise: (i) how to efficiently use pre-collected off-policy, non-optimal behavior data; and (ii) how to mediate among different competing objectives and constraints. We thus study the problem of batch policy learning under multiple constraints, and offer a systematic solution. We first propose a flexible meta-algorithm that admits any batch reinforcement learning and online learning procedure as subroutines. We then present a specific algorithmic instantiation and provide performance guarantees for the main objective and all constraints. To certify constraint satisfaction, we propose a new and simple method for off-policy policy evaluation (OPE) and derive PAC-style bounds. Our algorithm achieves strong empirical results in different domains, including in a challenging problem of simulated car driving subject to multiple constraints such as lane keeping and smooth driving. We also show experimentally that our OPE method outperforms other popular OPE techniques on a standalone basis, especially in a high-dimensional setting.

References in corpus (9)

• Trust Region Policy Optimization
• Reinforcement Learning with Deep Energy-Based Policies
• Deep Q-learning from Demonstrations
• Reinforcement and Imitation Learning via Interactive No-Regret Learning
• Breaking the Curse of Horizon: Infinite-Horizon Off-Policy Estimation
• Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic
• Transfer from Multiple MDPs
• Using Options and Covariance Testing for Long Horizon Off-Policy Policy Evaluation
• Smooth Imitation Learning for Online Sequence Prediction