mlsanityData to 14 Oct 2026

A universal policy wrapper with guarantees

arXiv 2505.12354 ↗First posted 18 May 2025cs.LG cs.AI cs.ROcited by 0

On the map: Deep reinforcement learning methods → Safe and constrained RL

Abstract (arXiv)

We introduce a universal policy wrapper for reinforcement learning agents that ensures formal goal-reaching guarantees. In contrast to standard reinforcement learning algorithms that excel in performance but lack rigorous safety assurances, our wrapper selectively switches between a high-performing base policy -- derived from any existing RL method -- and a fallback policy with known convergence properties. Base policy's value function supervises this switching process, determining when the fallback policy should override the base policy to ensure the system remains on a stable path. The analysis proves that our wrapper inherits the fallback policy's goal-reaching guarantees while preserving or improving upon the performance of the base policy. Notably, it operates without needing additional system knowledge or online constrained optimization, making it readily deployable across diverse reinforcement learning architectures and tasks.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Safe and constrained RL, one of 15 directions in Deep reinforcement learning methods. Builds on 13 of its 25 references that are in the map. First posted on arXiv 18 May 2025.

Builds on

13 of 25 references.

References · 13

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  3. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  4. 2017Constrained Policy Optimizationcited by 983
  5. 2019Solving Rubik's Cube with a Robot Handcited by 734
  6. 2017Safe Reinforcement Learning via Shieldingcited by 449
  7. 2018Reward Constrained Policy Optimizationcited by 417
  8. 2019Control Barrier Functions: Theory and Applicationscited by 372
  9. 2018A Lyapunov-based Approach to Safe Reinforcement Learningcited by 297
  10. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 289
  11. 2018Safe Exploration in Continuous Action Spacescited by 247
  12. 2022Constrained Update Projection Approach to Safe Policy Optimizationcited by 58
1 more
  1. 2024Critic as Lyapunov function (CALF): a model-free, stability-ensuring agentcited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024A novel agent with formal goal-reaching guarantees: an experimental study with a mobile robot
  2. 2026Reachability-Aware Diffusion Policy Optimization
  3. 2026A Survey on the Verification of Reinforcement Learning Policies
  4. 2025A Provable Approach for End-to-End Safe Reinforcement Learning
  5. 2026SafeExplorer: An Unbiased Policy Gradient for Reinforcement Learning with Recovery Interventions
  6. 2026Learning When to Act: Communication-Efficient Reinforcement Learning via Run-Time Assurance
  7. 2025R2L: Reliable Reinforcement Learning: Guaranteed Return & Reliable Policies in Reinforcement Learning
  8. 2025Making Universal Policies Universal