Reward Poisoning Attack Against Offline Reinforcement Learning

arXiv:2402.09695v1 Announce Type: new
Abstract: We study the problem of reward poisoning attacks against general offline reinforcement learning with deep neural networks for function approximation. We consider a black-box threat model where the attacker is completely oblivious to the learning algorithm and its budget is limited by constraining both the amount of corruption at each data point, and the total perturbation. We propose an attack strategy called `policy contrast attack'. The high-level idea is to make some low-performing policies appear …

abstract algorithm approximation arxiv attacks box budget corruption cs.ai cs.lg function general networks neural networks offline poisoning attacks reinforcement reinforcement learning study threat type