mlsanityData to 14 Oct 2026

BASIL: Best-Action Symbolic Interpretable Learning for Evolving Compact RL Policies

arXiv 2506.00328 ↗First posted 31 May 2025cs.AIcited by 0

On the map: Deep reinforcement learning methods → Interpretable and logic-guided RL

Abstract (arXiv)

The quest for interpretable reinforcement learning is a grand challenge for the deployment of autonomous decision-making systems in safety-critical applications. Modern deep reinforcement learning approaches, while powerful, tend to produce opaque policies that compromise verification, reduce transparency, and impede human oversight. To address this, we introduce BASIL (Best-Action Symbolic Interpretable Learning), a systematic approach for generating symbolic, rule-based policies via online evolutionary search with quality-diversity (QD) optimization. BASIL represents policies as ordered lists of symbolic predicates over state variables, ensuring full interpretability and tractable policy complexity. By using a QD archive, the methodology in the proposed study encourages behavioral and structural diversity between top-performing solutions, while a complexity-aware fitness encourages the synthesis of compact representations. The evolutionary system supports the use of exact constraints for rule count and system adaptability for balancing transparency with expressiveness. Empirical comparisons with three benchmark tasks CartPole-v1, MountainCar-v0, and Acrobot-v1 show that BASIL consistently synthesizes interpretable controllers with compact representations comparable to deep reinforcement learning baselines. Herein, this article introduces a new interpretable policy synthesis method that combines symbolic expressiveness, evolutionary diversity, and online learning through a unifying framework.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Interpretable and logic-guided RL, one of 15 directions in Deep reinforcement learning methods. Builds on 11 of its 26 references that are in the map. First posted on arXiv 31 May 2025.

Builds on

11 of 26 references.

References · 11

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  3. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  4. 2024Reinforcement learningcited by 247
  5. 2014Robots that can adapt like animalscited by 204
  6. 2018Programmatically Interpretable Reinforcement Learningcited by 193
  7. 2017Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learningcited by 150
  8. 2022Quality Diversity Evolutionary Learning of Decision Treescited by 7
  9. 2018Generating Interpretable Fuzzy Controllers using Particle Swarm Optimization and Genetic Programmingcited by 5
  10. 2021BF++: a language for general-purpose program synthesiscited by 3
  11. 2022Distilling Deep RL Models Into Interpretable Neuro-Fuzzy Systemscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Discovering Interpretable Multi-Parameter Control Policies for Evolutionary Algorithms Using Deep Reinforcement Learning
  2. 2025Surrogate Fitness Metrics for Interpretable Reinforcement Learning
  3. 2024Mitigating Information Loss in Tree-Based Reinforcement Learning via Direct Optimization
  4. 2026Interpretable reinforcement learning with decision-tree pruning
  5. 2024Distilling Reinforcement Learning Policies for Interpretable Robot Locomotion: Gradient Boosting Machines and Symbolic Regression
  6. 2026Code Evolution for Control: Synthesizing Policies via LLM-Driven Evolutionary Search
  7. 2024Towards a Research Community in Interpretable Reinforcement Learning: the InterpPol Workshop
  8. 2025Balancing Interpretability and Performance in Reinforcement Learning: An Adaptive Spectral Based Linear Approach