A Survey on Explainable Deep Reinforcement Learning

TL;DR

This survey reviews explainable deep reinforcement learning methods, their evaluation, and integration with large language models, aiming to improve transparency, trust, and safety in AI decision-making systems.

Contribution

It provides a comprehensive overview of XRL techniques, assessment frameworks, and explores the integration of RL with LLMs like RLHF for better AI alignment.

Findings

XRL enhances transparency at multiple levels

Evaluation frameworks for XRL are established

Integration of RL with LLMs improves AI alignment

Abstract

Deep Reinforcement Learning (DRL) has achieved remarkable success in sequential decision-making tasks across diverse domains, yet its reliance on black-box neural architectures hinders interpretability, trust, and deployment in high-stakes applications. Explainable Deep Reinforcement Learning (XRL) addresses these challenges by enhancing transparency through feature-level, state-level, dataset-level, and model-level explanation techniques. This survey provides a comprehensive review of XRL methods, evaluates their qualitative and quantitative assessment frameworks, and explores their role in policy refinement, adversarial robustness, and security. Additionally, we examine the integration of reinforcement learning with Large Language Models (LLMs), particularly through Reinforcement Learning from Human Feedback (RLHF), which optimizes AI alignment with human preferences. We conclude by highlighting open research challenges and future directions to advance the development of interpretable, reliable, and accountable DRL systems.