mlsanityData to 13 Oct 2026

Safety-Oriented Pruning and Interpretation of Reinforcement Learning Policies

arXiv 2409.10218 ↗First posted 16 Sep 2024cs.LGcited by 9

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

Abstract (arXiv)

Pruning neural networks (NNs) can streamline them but risks removing vital parameters from safe reinforcement learning (RL) policies. We introduce an interpretable RL method called VERINTER, which combines NN pruning with model checking to ensure interpretable RL safety. VERINTER exactly quantifies the effects of pruning and the impact of neural connections on complex safety properties by analyzing changes in safety measurements. This method maintains safety in pruned RL policies and enhances understanding of their safety dynamics, which has proven effective in multiple RL settings.

Key facts. Cited by 9 papers in the map (data through 13 Oct 2026). Placed in Safe and constrained RL, one of 15 directions in Deep reinforcement learning methods. Builds on 9 of its 18 references that are in the map. First posted on arXiv 16 Sep 2024.

Builds on

9 of 18 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Probabilistic Model Checking of Stochastic Reinforcement Learning Policies27 Mar

Other references · 8

  1. 2013Playing Atari with Deep Reinforcement Learningcited by 3,221
  2. 2022A Survey of Multi-Agent Deep Reinforcement Learning with Communicationcited by 104
  3. 2018Omega-Regular Objectives in Model-Free Reinforcement Learningcited by 74
  4. 2020The Probabilistic Model Checker Stormcited by 64
  5. 2021Scalar reward is not enough: A response to Silver, Singh, Precup and Sutton (2021)cited by 55
  6. 2014Verification of Markov Decision Processes using Learning Algorithmscited by 38
  7. 2022COOL-MC: A Comprehensive Tool for Reinforcement Learning and Model Checkingcited by 19
  8. 2020Statistically Model Checking PCTL Specifications on Markov Decision Processes via Reinforcement Learningcited by 12

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 4; 2026: 5.

9 papers.

  1. 2026Formally Verifying and Explaining Sepsis Treatment Policies with COOL-MCcited by 3
  2. 2025Co-Activation Graph Analysis of Safety-Verified and Explainable Deep Reinforcement Learning Policiescited by 2
  3. 2025Verifying Memoryless Sequential Decision-making of Large Language Modelscited by 1
  4. 2025Formal Verification of Noisy Quantum Reinforcement Learning Policiescited by 0
  5. 2025Translating the Rashomon Effect to Sequential Decision-Making Taskscited by 0
  6. 2026COOL-MC: Verifying and Explaining RL Policies for Platelet Inventory Managementcited by 0
  7. 2026COOL-MC: Verifying and Explaining RL Policies for Multi-bridge Network Maintenancecited by 0
  8. 2026Formal Verification of Learned Multi-Agent Communication Policies via Decision Tree Distillationcited by 0
  9. 2026Probabilistic Model Checking of Autoregressive Neural Sequence Modelscited by 0