AI & ML
Safe Reinforcement Learning with Hard Constraints: Policy Gradient Methods with Intersample Guarantees
arXiv cs.SY0 views
arXiv:2611.21987v1 Announce Type: new Abstract: We develop a policy-gradient (PG) reinforcement learning (RL) framework with hard inter-sample control barrier function (CBF)-based safety constraints for nonlinear sampled-data systems. We first design a class of stochastic policies with support truncated to sampling-period-dependent safe control sets, that guarantee safety-by-construction at sampling instants. We then propose a model-free PG algorithm that searches over the class of safe-by-design policies and establish sufficient conditions on sampling-period such that any action sampled from safe control sets remains safe throughout the sampling interval, providing inter-sample guarantees. We further establish finite-time convergence and sample complexity guarantees for the proposed PG algorithm and characterize its global optimality under gradient dominance. Finally, we illustrate inter-sample safety guarantees on a safe quadrotor navigation environment.
Source:https://arxiv.org/abs/2611.21987PreviousFirst-Order Model-Free Policy Gradient for Undiscounted LQR with Global Optimality and Convergence GuaranteesNextPS-HalluBench: Taxonomy-Driven Benchmarking of LLM Hallucinations in Power Systems
Related posts
My context grew 6x in three months. That decides the cache question
0dev.to AI
How to connect Joist to any MCP client: Claude, ChatGPT, Cursor
0dev.to AI
Gemini Notebook cria sites interativos a partir de anotações
0dev.to AI
Two customers bought the same box of screws within five minutes and SQS delivered one sale
0dev.to AWS
AI Visibility in 2026: The Human–Agent Playbook for SEO, AEO & GEO
0dev.to WebDev
How to Trim Audio for a Beekeeping Tutorial Video — Free, in Your Browser
0dev.to WebDev