Verified Safe Reinforcement Learning for Neural Network Dynamic Models

2024-05-25 · Junlin Wu, huan zhang, Yevgeniy Vorobeychik

Learning reliably safe autonomous control is one of the core problems in trustworthy autonomy. However, training a controller that can be formally verified to be safe remains a major challenge. We introduce a novel approach for learning verified safe control policies in nonlinear neural dynamical systems while maximizing overall performance. Our approach aims to achieve safety in the sense of finite-horizon reachability proofs, and is comprised of three key parts. The first is a novel curriculum learning scheme that iteratively increases the verified safe horizon. The second leverages the iterative nature of gradient-based learning to leverage incremental verification, reusing information from prior verification runs. Finally, we learn multiple verified initial-state-dependent controllers, an idea that is especially valuable for more complex domains where learning a single universal verified safe controller is extremely challenging. Our experiments on five safe control problems demonstrate that our trained controllers can achieve verified safety over horizons that are as much as an order of magnitude longer than state-of-the-art baselines, while maintaining high reward, as well as a perfect safety record over entire episodes. Our code is available at https://github.com/jlwu002/VSRL.

📄 PDF Abstract BibTeX arXiv:2405.15994

Code (1)

jlwu002/vsrl 공식 구현 pytorch

Tasks

reinforcement-learning Reinforcement LearningSafe Reinforcement Learning

Runtime Safety Assurance Using Reinforcement Learning

2020-10-20 · Christopher Lazarus, James G. Lopez, Mykel J. Kochenderfer

The airworthiness and safety of a non-pedigreed autopilot must be verified, but the cost to formally do so can be prohibitive. We can bypass formal verification of non-pedigreed components by incorporating Runtime Safety…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Provably Safe Deep Reinforcement Learning for Robotic Manipulation in Human Environments

2022-05-12 · Jakob Thumm, Matthias Althoff

Deep reinforcement learning (RL) has shown promising results in the motion planning of manipulators. However, no method guarantees the safety of highly dynamic obstacles, such as humans, in RL-based manipulator control. …

Deep Reinforcement Learning Motion Planning reinforcement-learning Reinforcement Learning (RL)

Joint Synthesis of Safety Certificate and Safe Control Policy using Constrained Reinforcement Learning

2021-11-15 · Haitong Ma, Changliu Liu, Shengbo Eben Li, Sifa Zheng 외

Safety is the major consideration in controlling complex dynamical systems using reinforcement learning (RL), where the safety certificate can provide provable safety guarantee. A valid safety certificate is an energy fu…

reinforcement-learning Reinforcement Learning (RL)valid

Safe Reinforcement Learning using Ideas from Model Predictive Control

2026-07-08 · Georg Schäfer, Jakob Rehrl, Stefan Huber, Simon Hirlaender arxiv

Reinforcement learning (RL) enables the synthesis of control policies directly from data, making it highly appealing for complex cyber-physical systems (CPSs) and robotics. A persistent challenge, however, is ensuring st…

Reinforcement Learning Active Learning

Neurosymbolic Reinforcement Learning with Formally Verified Exploration

2020-09-26 · NeurIPS 2020 12 · Greg Anderson, Abhinav Verma, Isil Dillig, Swarat Chaudhuri

We present Revel, a partially neural reinforcement learning (RL) framework for provably safe exploration in continuous state and action spaces. A key challenge for provably safe deep RL is that repeatedly verifying neura…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)Safe Exploration