Adaptive Shielding for Safe Reinforcement Learning under Hidden-Parameter Dynamics Shifts

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kwon, Minjae, Ingebrand, Tyler, Topcu, Ufuk, Feng, Lu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908800828571648
author Kwon, Minjae
Ingebrand, Tyler
Topcu, Ufuk
Feng, Lu
author_facet Kwon, Minjae
Ingebrand, Tyler
Topcu, Ufuk
Feng, Lu
contents Unseen shifts in environment dynamics, driven by hidden parameters such as friction or gravity, create a challenge for maintaining safety. We address this challenge by proposing Adaptive Shielding, a framework for safe reinforcement learning in constrained hidden-parameter Markov decision processes. A function encoder infers a low-dimensional representation of the underlying dynamics online from transition data, allowing the shield to adapt. To ensure safety during this process, we use a two-layer strategy. First, we introduce safety-regularized optimization that proactively trains the policy away from high-cost regions. Second, the adaptive shielding reactively uses the inferred dynamics to forecast safety risks and applies uncertainty-aware bounds using conformal prediction to filter unsafe actions. We prove that prediction errors in the shielding connect with bounds on the average cost rate. Empirically, across Safe-Gym benchmarks with varying hidden parameters, our approach outperforms baselines on the return-safety trade-off and generalizes reliably to unseen dynamics, while incurring only modest execution-time overhead. Code is available at https://github.com/safe-autonomy-lab/AdaptiveShieldingFE.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11033
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Shielding for Safe Reinforcement Learning under Hidden-Parameter Dynamics Shifts
Kwon, Minjae
Ingebrand, Tyler
Topcu, Ufuk
Feng, Lu
Machine Learning
Artificial Intelligence
Unseen shifts in environment dynamics, driven by hidden parameters such as friction or gravity, create a challenge for maintaining safety. We address this challenge by proposing Adaptive Shielding, a framework for safe reinforcement learning in constrained hidden-parameter Markov decision processes. A function encoder infers a low-dimensional representation of the underlying dynamics online from transition data, allowing the shield to adapt. To ensure safety during this process, we use a two-layer strategy. First, we introduce safety-regularized optimization that proactively trains the policy away from high-cost regions. Second, the adaptive shielding reactively uses the inferred dynamics to forecast safety risks and applies uncertainty-aware bounds using conformal prediction to filter unsafe actions. We prove that prediction errors in the shielding connect with bounds on the average cost rate. Empirically, across Safe-Gym benchmarks with varying hidden parameters, our approach outperforms baselines on the return-safety trade-off and generalizes reliably to unseen dynamics, while incurring only modest execution-time overhead. Code is available at https://github.com/safe-autonomy-lab/AdaptiveShieldingFE.
title Adaptive Shielding for Safe Reinforcement Learning under Hidden-Parameter Dynamics Shifts
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.11033