Safe Reinforcement Learning with Learned Non-Markovian Safety Constraints

TL;DR

This paper introduces a scalable reinforcement learning framework that learns to satisfy complex non-Markovian safety constraints by modeling safety as a trajectory-based credit assignment problem and dynamically balancing safety and reward.

Contribution

It proposes a novel safety model for credit assignment on trajectories, an RL-as-inference algorithm for safe policy optimization, and a dynamic method to balance safety and reward.

Findings

Successfully handles non-Markovian safety constraints

Scalable approach demonstrated on complex safety tasks

Effectively balances safety and reward during training

Abstract

In safe Reinforcement Learning (RL), safety cost is typically defined as a function dependent on the immediate state and actions. In practice, safety constraints can often be non-Markovian due to the insufficient fidelity of state representation, and safety cost may not be known. We therefore address a general setting where safety labels (e.g., safe or unsafe) are associated with state-action trajectories. Our key contributions are: first, we design a safety model that specifically performs credit assignment to assess contributions of partial state-action trajectories on safety. This safety model is trained using a labeled safety dataset. Second, using RL-as-inference strategy we derive an effective algorithm for optimizing a safe policy using the learned safety model. Finally, we devise a method to dynamically adapt the tradeoff coefficient between reward maximization and safety compliance. We rewrite the constrained optimization problem into its dual problem and derive a gradient-based method to dynamically adjust the tradeoff coefficient during training. Our empirical results demonstrate that this approach is highly scalable and able to satisfy sophisticated non-Markovian safety constraints.