# Verifiable Reinforcement Learning via Policy Extraction

arXiv: https://arxiv.org/abs/1805.08328 · HTML: https://mlsanity.com/paper/1805.08328/

## Key facts

- Cited by 177 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 May 2018.

## Built on by (most cited first, 30 of 177)

- [Scalable agent alignment via reward modeling: a research direction](https://mlsanity.com/paper/1811.07871/) (2018)
- [Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges](https://mlsanity.com/paper/2103.11251/) (2021)
- [Multi-Game Decision Transformers](https://mlsanity.com/paper/2205.15241/) (2022)
- [A Review of Safe Reinforcement Learning: Methods, Theory and Applications](https://mlsanity.com/paper/2205.10330/) (2022)
- [A Survey on Interpretable Reinforcement Learning](https://mlsanity.com/paper/2112.13112/) (2021)
- [Learning to Synthesize Programs as Interpretable and Generalizable Policies](https://mlsanity.com/paper/2108.13643/) (2021)
- [Provable Repair of Deep Neural Networks](https://mlsanity.com/paper/2104.04413/) (2021)
- [Safe Reinforcement Learning via Probabilistic Logic Shields](https://mlsanity.com/paper/2303.03226/) (2023)
- [Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZero](https://mlsanity.com/paper/2310.16410/) (2023)
- [Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search](https://mlsanity.com/paper/2405.15383/) (2024)
- [On Transforming Reinforcement Learning by Transformer: The Development Trajectory](https://mlsanity.com/paper/2212.14164/) (2022)
- [Provably Safe Reinforcement Learning: Conceptual Analysis, Survey, and Benchmarking](https://mlsanity.com/paper/2205.06750/) (2022)
- [A Review of Formal Methods applied to Machine Learning](https://mlsanity.com/paper/2104.02466/) (2021)
- [Explainable Deep Reinforcement Learning: State of the Art and Challenges](https://mlsanity.com/paper/2301.09937/) (2023)
- [RL-GPT: Integrating Reinforcement Learning and Code-as-policy](https://mlsanity.com/paper/2402.19299/) (2024)
- [Iterative Bounding MDPs: Learning Interpretable Policies via Non-Interpretable Methods](https://mlsanity.com/paper/2102.13045/) (2021)
- [A Survey of Explainable Reinforcement Learning](https://mlsanity.com/paper/2202.08434/) (2022)
- [HardNet: Hard-Constrained Neural Networks with Universal Approximation Guarantees](https://mlsanity.com/paper/2410.10807/) (2024)
- [Neurosymbolic Reinforcement Learning and Planning: A Survey](https://mlsanity.com/paper/2309.01038/) (2023)
- [Interpretable Concept Bottlenecks to Align Reinforcement Learning Agents](https://mlsanity.com/paper/2401.05821/) (2024)
- [ProTo: Program-Guided Transformer for Program-Guided Tasks](https://mlsanity.com/paper/2110.00804/) (2021)
- [Hierarchical Programmatic Reinforcement Learning via Learning to Compose Programs](https://mlsanity.com/paper/2301.12950/) (2023)
- [Interpretable and Editable Programmatic Tree Policies for Reinforcement Learning](https://mlsanity.com/paper/2405.14956/) (2024)
- [GALOIS: Boosting Deep Reinforcement Learning via Generalizable Logic Synthesis](https://mlsanity.com/paper/2205.13728/) (2022)
- [Dynamic Model Predictive Shielding for Provably Safe Reinforcement Learning](https://mlsanity.com/paper/2405.13863/) (2024)
- [Neurosymbolic Transformers for Multi-Agent Communication](https://mlsanity.com/paper/2101.03238/) (2021)
- [Synthesizing Programmatic Reinforcement Learning Policies with Large Language Model Guided Search](https://mlsanity.com/paper/2405.16450/) (2024)
- [Mo"ET: Mixture of Expert Trees and its Application to Verifiable Reinforcement Learning](https://mlsanity.com/paper/1906.06717/) (2019)
- [Improving Human Sequential Decision-Making with Reinforcement Learning](https://mlsanity.com/paper/2108.08454/) (2021)
- [MAVIPER: Learning Decision Tree Policies for Interpretable Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2205.12449/) (2022)

## Abstract (arXiv)

While deep reinforcement learning has successfully solved many challenging control tasks, its real-world applicability has been limited by the inability to ensure the safety of learned policies. We propose an approach to verifiable reinforcement learning by training decision tree policies, which can represent complex policies (since they are nonparametric), yet can be efficiently verified using existing techniques (since they are highly structured). The challenge is that decision tree policies are difficult to train. We propose VIPER, an algorithm that combines ideas from model compression and imitation learning to learn decision tree policies guided by a DNN policy (called the oracle) and its Q-function, and show that it substantially outperforms two baselines. We use VIPER to (i) learn a provably robust decision tree policy for a variant of Atari Pong with a symbolic state space, (ii) learn a decision tree policy for a toy game based on Pong that provably never loses, and (iii) learn a provably stable decision tree policy for cart-pole. In each case, the decision tree policy achieves performance equal to that of the original DNN policy.