# Interpretable and logic-guided RL

Area: [Deep reinforcement learning methods](https://mlsanity.com/deeprl/index.md). HTML: https://mlsanity.com/deeprl/interpretable-and-logic-guided-rl/

## Key facts

- Interpretable and logic-guided RL: 195 papers on arXiv since 2024 (data through 14 Oct 2026).
- 13 of its 195 mapped papers (7%) cite “DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RL” (2024, arXiv 2410.04631).
- 12 of its 195 mapped papers (6%) cite “Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learning” (2024, arXiv 2411.00205).
- 3 new papers 30 Sep–14 Oct.

## Most cited here

- [DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RL](https://mlsanity.com/paper/2410.04631/) (2024), cited by 13 papers here
- [Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learning](https://mlsanity.com/paper/2411.00205/) (2024), cited by 12 papers here
- [Interpretable Concept Bottlenecks to Align Reinforcement Learning Agents](https://mlsanity.com/paper/2401.05821/) (2024), cited by 11 papers here
- [LTL-Constrained Policy Optimization with Cycle Experience Replay](https://mlsanity.com/paper/2404.11578/) (2024), cited by 8 papers here
- [End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanations](https://mlsanity.com/paper/2403.12451/) (2024), cited by 7 papers here
- [Reinforcement Learning with LTL and ω-Regular Objectives via Optimality-Preserving Translation to Average Rewards](https://mlsanity.com/paper/2410.12175/) (2024), cited by 7 papers here
- [Interpretable and Editable Programmatic Tree Policies for Reinforcement Learning](https://mlsanity.com/paper/2405.14956/) (2024), cited by 6 papers here
- [Reward Machines for Deep RL in Noisy and Uncertain Environments](https://mlsanity.com/paper/2406.00120/) (2024), cited by 6 papers here
- [Directed Exploration in Reinforcement Learning from Linear Temporal Logic](https://mlsanity.com/paper/2408.09495/) (2024), cited by 6 papers here
- [One Subgoal at a Time: Zero-Shot Generalization to Arbitrary Linear Temporal Logic Requirements in Multi-Task Reinforcement Learning](https://mlsanity.com/paper/2508.01561/) (2025), cited by 6 papers here

## New (30 Sep–14 Oct)

- [Visualizing the World Model in an Agent's Value Function](https://mlsanity.com/paper/2610.14186/) (2026-10-09)
- [Towards the Automatic Synthesis of Interpretable Chess Tactics](https://mlsanity.com/paper/2610.07640/) (2026-10-06)
- [Minimal Witness Reinforcement Learning](https://mlsanity.com/paper/2610.07226/) (2026-10-05)