Timezone: »

Adversarial Attacks on Linear Contextual Bandits
Evrard Garcelon · Baptiste Roziere · Laurent Meunier · Jean Tarbouriech · Olivier Teytaud · Alessandro Lazaric · Matteo Pirotta

Tue Dec 08 09:00 AM -- 11:00 AM (PST) @ Poster Session 1 #244

Contextual bandit algorithms are applied in a wide range of domains, from advertising to recommender systems, from clinical trials to education. In many of these domains, malicious agents may have incentives to force a bandit algorithm into a desired behavior For instance, an unscrupulous ad publisher may try to increase their own revenue at the expense of the advertisers; a seller may want to increase the exposure of their products, or thwart a competitor’s advertising campaign. In this paper, we study several attack scenarios and show that a malicious agent can force a linear contextual bandit algorithm to pull any desired arm T − o(T) times over a horizon of T steps, while applying adversarial modifications to either rewards or contexts with a cumulative cost that only grow logarithmically as O(log T). We also investigate the case when a malicious agent is interested in affecting the behavior of the bandit algorithm in a single context (e.g., a specific user). We first provide sufficient conditions for the feasibility of the attack and an efficient algorithm to perform an attack. We empirically validate the proposed approaches on synthetic and real-world datasets.

Author Information

Evrard Garcelon (Facebook AI Research)
Baptiste Roziere (Facebook AI Research and Paris-Dauphine University)
Laurent Meunier (Dauphine University - FAIR Paris)
Jean Tarbouriech (Facebook AI Research & Inria)
Olivier Teytaud (Facebook)
Alessandro Lazaric (Facebook Artificial Intelligence Research)
Matteo Pirotta (Facebook AI Research)

More from the Same Authors