Universal Trojan Signatures in Reinforcement Learning

Manoj Acharya, Weichao Zhou, Anirban Roy, Xiao Lin, Wenchao Li, Susmit Jha

Keywords: Trojan, RL, reinforcement learning, backdoor attacks, attribution, Jacobian

TL;DR: We use attribution analysis to detect trojan models which can generalize to transfer settings with novel RL environments and modified architectures.

Abstract: We present a novel approach for characterizing Trojaned reinforcement learning (RL) agents. By monitoring for discrepancies in how an agent's policy evaluates state observations for choosing an action, we can reliably detect whether the policy is Trojaned. Experiments on the IARPA RL challenge benchmarks show that our approach can effectively detect Trojaned models even in transfer settings with novel RL environments and modified architectures.

Submission Number: 34