Safe Reinforcement Learning by Imagining the Near Future

Garrett Thomas, Yuping Luo, Tengyu Ma

Keywords: safe reinforcement learning, model-based reinforcement learning

TL;DR: We devise a short-horizon model-based reinforcement learning algorithm to avoid safety violations by penalizing unsafe trajectories.

Abstract:

Safe reinforcement learning is a promising path toward applying reinforcement learning algorithms to real-world problems, where suboptimal behaviors may lead to actual negative consequences. In this work, we focus on the setting where unsafe states can be avoided by planning ahead a short time into the future. In this setting, a model-based agent with a sufficiently accurate model can avoid unsafe states.
We devise a model-based algorithm that heavily penalizes unsafe trajectories, and derive guarantees that our algorithm can avoid unsafe states under certain assumptions. Experiments demonstrate that our algorithm can achieve competitive rewards with fewer safety violations in several continuous control tasks.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/gwthomas/Safe-MBPO

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:2202.07789/code)