Efficient Action Poisoning Attacks on Linear Contextual Bandits
Abstract
Contextual bandit algorithms have many applicants in a variety of scenarios. In order to develop trustworthy contextual bandit systems, understanding the impacts of various adversarial attacks on contextual bandit algorithms is essential. In this paper, we propose a new class of attacks: action poisoning attacks, where an adversary can change the action signal selected by the agent. We design action poisoning attack schemes against linear contextual bandit algorithms in both white-box and black-box settings. We further analyze the cost of the proposed attack strategies for a very popular and widely used bandit algorithm: LinUCB. We show that, in both white-box and black-box settings, the proposed attack schemes can force the LinUCB agent to pull a target arm very frequently by spending only logarithm cost.
Keywords
Cite
@article{arxiv.2112.05367,
title = {Efficient Action Poisoning Attacks on Linear Contextual Bandits},
author = {Guanlin Liu and Lifeng Lai},
journal= {arXiv preprint arXiv:2112.05367},
year = {2021}
}