# BARD-MARL: Byzantine-Agent Detection for Learned Communication in Multi-Agent Reinforcement Learning

arXiv: https://arxiv.org/abs/2606.20701 · HTML: https://mlsanity.com/paper/2606.20701/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 15 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Builds on (most cited first, 10 of 10)

- [Verifiable Reinforcement Learning via Policy Extraction](https://mlsanity.com/paper/1805.08328/) (2018)
- [A Survey of Multi-Agent Deep Reinforcement Learning with Communication](https://mlsanity.com/paper/2203.08975/) (2022)
- [Generation of Policy-Level Explanations for Reinforcement Learning](https://mlsanity.com/paper/1905.12044/) (2019)
- [Byzantine Robust Cooperative Multi-Agent Reinforcement Learning as a Bayesian Game](https://mlsanity.com/paper/2305.12872/) (2023)
- [Towards Resilience for Multi-Agent QD-Learning](https://mlsanity.com/paper/2104.03153/) (2021)
- [Bayesian Ego-graph Inference for Networked Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2509.16606/) (2025)
- [Decentralized Vision-Based Byzantine Agent Detection in Multi-Robot Systems with IOTA Smart Contracts](https://mlsanity.com/paper/2210.03441/) (2022)
- [On the Hardness of Decentralized Multi-Agent Policy Evaluation under Byzantine Attacks](https://mlsanity.com/paper/2409.12882/) (2024)
- [Reinforcement Learning based Cyberattack Model for Adaptive Traffic Signal Controller in Connected Transportation Systems](https://mlsanity.com/paper/2211.01845/) (2022)
- [siForest: Detecting Network Anomalies with Set-Structured Isolation Forest](https://mlsanity.com/paper/2412.06015/) (2024)

## Abstract (arXiv)

Learned communication improves coordination in cooperative multi-agent reinforcement learning, but it also creates a trust problem: a trained policy may route information through agents that have become faulty or adversarial. This paper studies Byzantine-agent detection for learned-communication MARL in adaptive traffic signal control. We propose BARD-MARL, a post-hoc diagnostic layer on top of BayesG, which is used as an attributed communication substrate rather than as a contribution of this paper. BARD-MARL combines two agent-level evidence streams: policy-graph features extracted from state-action trajectories and Bayesian trust statistics computed from BayesG latent mask probabilities. Across fixed-action, observation-flip, random-noise, and coordinated attacks in SUMO traffic grids, the results show that these signals are complementary rather than universally dominant. On a 25-agent grid, BARD-MARL reaches 0.843 AUC-ROC under a 10% observation-flip attack, while policy-graph-only detection reaches 0.917 AUC-ROC under a 10% coordinated attack. On a 100-agent grid, the unified BARD-MARL variant reaches 0.982 AUC-ROC for both 10% fixed-action and 10% coordinated attacks. The study shows that learned communication policies expose useful diagnostic evidence, but credible resilience claims require attack-specific ablations and explicit separation between coordination, detection, and mitigation.