mlsanityData to 7 Oct 2026

Verified Probabilistic Policies for Deep Reinforcement Learning

arXiv 2201.03698 ↗First posted 10 Jan 2022cs.AI cs.LG cs.LOcited by 6

Not on the map. Why?

Abstract (arXiv)

Deep reinforcement learning is an increasingly popular technique for synthesising policies to control an agent's interaction with its environment. There is also growing interest in formally verifying that such policies are correct and execute safely. Progress has been made in this area by building on existing work for verification of deep neural networks and of continuous-state dynamical systems. In this paper, we tackle the problem of verifying probabilistic policies for deep reinforcement learning, which are used to, for example, tackle adversarial environments, break symmetries and manage trade-offs. We propose an abstraction approach, based on interval Markov decision processes, that yields probabilistic guarantees on a policy's execution, and present techniques to build and solve these models using abstract interpretation, mixed-integer linear programming, entropy-based refinement and probabilistic model checking. We implement our approach and illustrate its effectiveness on a selection of reinforcement learning benchmarks.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 21 of its 50 references that are in the map. First posted on arXiv 10 Jan 2022.

Builds on

21 of 50 references.

References · 21

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,218
  3. 2017Safe Reinforcement Learning via Shieldingcited by 427
  4. 2016Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updatescited by 342
  5. 2019Reinforcement Learning in Healthcare: A Surveycited by 308
  6. 2018Learning to Drive in a Daycited by 266
  7. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 256
  8. 2018Verifiable Reinforcement Learning via Policy Extractioncited by 176
  9. 2020Learning to be Safe: Deep RL with a Safety Criticcited by 109
  10. 2017Maximum Resilience of Artificial Neural Networkscited by 97
  11. 2020Agent Modelling under Partial Observability for Deep Reinforcement Learningcited by 56
  12. 2020Cautious Reinforcement Learning with Logical Constraintscited by 44
9 more
  1. 2019An Inductive Synthesis Framework for Verifiable Reinforcement Learningcited by 39
  2. 2020Verifiably Safe Exploration for End-to-End Reinforcement Learningcited by 32
  3. 2021Feasible Actor-Critic: Constrained Reinforcement Learning for Ensuring Statewise Safetycited by 30
  4. 2018Logically-Constrained Neural Fitted Q-Iterationcited by 17
  5. 2020Probabilistic Guarantees for Safe Deep Reinforcement Learningcited by 15
  6. 2019Efficiency through Uncertainty: Scalable Formal Synthesis for Stochastic Hybrid Systemscited by 11
  7. 2021Distillation of RL Policies with Formal Guarantees via Variational Abstraction of Markov Decision Processes (Technical Report)cited by 9
  8. 2018Approximate Abstractions of Markov Chains with Interval Decision Processes (Extended Version)cited by 4
  9. 2021Learning on Abstract Domains: A New Approach for Verifiable Guarantee in Reinforcement Learningcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 0; 2024: 1; 2025: 1; 2026: 3.

6 papers.

  1. 2022COOL-MC: A Comprehensive Tool for Reinforcement Learning and Model Checkingcited by 19
  2. 2024Unifying Qualitative and Quantitative Safety Verification of DNN-Controlled Systemscited by 6
  3. 2025A Dynamical Systems Framework for Reinforcement Learning Safety and Robustness Verificationcited by 0
  4. 2026COOL-MC: Verifying and Explaining RL Policies for Platelet Inventory Managementcited by 0
  5. 2026COOL-MC: Verifying and Explaining RL Policies for Multi-bridge Network Maintenancecited by 0
  6. 2026Automated Testing of LLM-Based Post Hoc Explainers Using Model Checking as an Oraclecited by 0