---
title: "BehaviorGuard: Online Backdoor Defense for Deep Reinforcement Learning"
authors:
  - "Yinbo Yu"
  - "Xueyu Yin"
  - "Jiadai Wang"
  - "Chunwei Tian"
  - "Sai Xu"
  - "Qi Zhu"
  - "Daoqiang Zhang"
doi: "10.24963/ijcai.2026/584"
url: "https://ijcai.org/proceedings/2026/584"
pdf: "https://ijcai.org/proceedings/2026/0584.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5245-5253"
track: "Main Track"
keywords:
  - "Multidisciplinary Topics and Applications: Security and privacy"
  - "Machine Learning: Reinforcement learning"
  - "Machine Learning: Multiagent Reinforcement Learning"
---

# BehaviorGuard: Online Backdoor Defense for Deep Reinforcement Learning

Yinbo Yu, Xueyu Yin, Jiadai Wang, Chunwei Tian, Sai Xu, Qi Zhu, Daoqiang Zhang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5245-5253. DOI: [10.24963/ijcai.2026/584](https://doi.org/10.24963/ijcai.2026/584)

## Abstract

Backdoor attacks pose a serious threat to deep reinforcement learning (DRL). Current defenses typically rely on reward anomalies to reverse-engineer triggers and model finetuning to remove backdoors. However, complex trigger patterns undermine their robustness, and fine-tuning entails high costs, limiting practical utility. To this end, we shift defense concerns to trigger-agnostic backdoor output behaviors and propose BehaviorGuard, an online behavior-based backdoor detection and mitigation framework for DRL. Specifically, we find that regardless of attacks, backdoored policies induce consistent shifts in action distributions to ensure reliable activation, leaving detectable traces in high-quantile regions and distribution tails, even in the absence of triggers. Based on this, we design a novel metric that captures behavioral drift in action distributions to identify and suppress backdoor actions at runtime. To our knowledge, this is the first online backdoor defense that counters attacks both in single- and multi-agent DRL. Evaluated across diverse benchmarks with different backdoor attacks, BehaviorGuard consistently surpasses prior methods in both efficacy and efficiency.