Unveiling the Black Box: A Multi-Layer Framework for Explaining Reinforcement Learning-Based Cyber Agents

TL;DR

This paper introduces a multi-layer explainability framework for reinforcement learning cyber agents, revealing strategic and tactical reasoning to improve trust, debugging, and threat understanding in cybersecurity.

Contribution

It presents a unified, domain-agnostic explainability approach that models cyberattacks as POMDPs and analyzes policy evolution, surpassing prior post-hoc methods.

Findings

Framework provides interpretable insights across increasing complexity environments.

Reveals exploration-exploitation dynamics and behavioral shifts.

Supports use cases like threat modeling and RL policy debugging.

Abstract

Reinforcement Learning (RL) agents are increasingly used to simulate sophisticated cyberattacks, but their decision-making processes remain opaque, hindering trust, debugging, and defensive preparedness. In high-stakes cybersecurity contexts, explainability is essential for understanding how adversarial strategies are formed and evolve over time. In this paper, we propose a unified, multi-layer explainability framework for RL-based attacker agents that reveals both strategic (Markov Decision Process (MDP)-level) and tactical (policy-level) reasoning. At the MDP-level, we model cyberattacks as a Partially Observable Markov Decision Process (POMDP) to expose exploration-exploitation dynamics and phase-aware behavioural shifts. At the policy-level, we analyse the temporal evolution of Q-values and use Prioritised Experience Replay (PER) to surface critical learning transitions and evolving action preferences. Evaluated across CyberBattleSim environments of increasing complexity, our framework offers interpretable insights into agent behaviour at scale. Unlike previous explainable RL methods, which are {predominantly} post-hoc, domain-specific, or limited in depth, our approach is both agent- and environment-agnostic, {supporting use cases such as red-team simulation, RL policy debugging, phase-aware threat modelling and anticipatory defence planning.} By transforming black-box learning into actionable behavioural intelligence, our framework enables both defenders and developers to better anticipate, analyse, and respond to autonomous cyber threats.