Timezone: »
Poster
Observation-Free Attacks on Stochastic Bandits
Yinglun Xu · Bhuvesh Kumar · Jacob Abernethy
We study data corruption attacks on stochastic multi arm bandit algorithms. Existing attack methodologies assume that the attacker can observe the multi arm bandit algorithm's realized behavior which is in contrast to the adversaries modeled in the robust multi arm bandit algorithms literature. To the best of our knowledge, we develop the first data corruption attack on stochastic multi arm bandit algorithms which works without observing the algorithm's realized behavior. Through this attack, we also discover a sufficient condition for a stochastic multi arm bandit algorithm to be susceptible to adversarial data corruptions. We show that any bandit algorithm that makes decisions just using the empirical mean reward, and the number of times that arm has been pulled in the past can suffer from linear regret under data corruption attacks. We further show that various popular stochastic multi arm bandit algorithms such UCB, $\epsilon$-greedy and Thompson Sampling satisfy this sufficient condition and are thus prone to data corruption attacks. We further analyze the behavior of our attack for these algorithms and show that using only $o(T)$ corruptions, our attack can force these algorithms to select a potentially non-optimal target arm preferred by the attacker for all but $o(T)$ rounds.
Author Information
Yinglun Xu (University of Illinois, Urbana Champaign)
Bhuvesh Kumar (Georgia Tech)
Jacob Abernethy (Georgia Institute of Technology)
More from the Same Authors
-
2019 Poster: Online Learning via the Differential Privacy Lens »
Jacob Abernethy · Young H Jung · Chansoo Lee · Audra McMillan · Ambuj Tewari -
2019 Spotlight: Online Learning via the Differential Privacy Lens »
Jacob Abernethy · Young H Jung · Chansoo Lee · Audra McMillan · Ambuj Tewari -
2019 Poster: Learning Auctions with Robust Incentive Guarantees »
Jacob Abernethy · Rachel Cummings · Bhuvesh Kumar · Sam Taggart · Jamie Morgenstern -
2018 : Panel discussion: Opportunities to organize new impactful challenges. »
Jacob Abernethy -
2018 : Panel discussion: Opportunities to organize new impactful challenges »
Jacob Abernethy -
2018 Workshop: CiML 2018 - Machine Learning competitions "in the wild": Playing in the real world or in real time »
Isabelle Guyon · Evelyne Viegas · Sergio Escalera · Jacob D Abernethy -
2018 : Building Algorithms by Playing Games »
Jacob D Abernethy -
2018 Poster: Acceleration through Optimistic No-Regret Dynamics »
Jun-Kun Wang · Jacob Abernethy -
2018 Spotlight: Acceleration through Optimistic No-Regret Dynamics »
Jun-Kun Wang · Jacob Abernethy -
2017 Workshop: Machine Learning Challenges as a Research Tool »
Isabelle Guyon · Evelyne Viegas · Sergio Escalera · Jacob D Abernethy -
2017 Poster: On Frank-Wolfe and Equilibrium Computation »
Jacob D Abernethy · Jun-Kun Wang -
2017 Spotlight: On Frank-Wolfe and Equilibrium Computation »
Jacob D Abernethy · Jun-Kun Wang -
2016 Poster: Threshold Bandits, With and Without Censored Feedback »
Jacob D Abernethy · Kareem Amin · Ruihao Zhu -
2015 Poster: Fighting Bandits with a New Kind of Smoothness »
Jacob D Abernethy · Chansoo Lee · Ambuj Tewari -
2015 Poster: A Market Framework for Eliciting Private Data »
Bo Waggoner · Rafael Frongillo · Jacob D Abernethy -
2014 Workshop: NIPS Workshop on Transactional Machine Learning and E-Commerce »
David Parkes · David H Wolpert · Jennifer Wortman Vaughan · Jacob D Abernethy · Amos Storkey · Mark Reid · Ping Jin · Nihar Bhadresh Shah · Mehryar Mohri · Luis E Ortiz · Robin Hanson · Aaron Roth · Satyen Kale · Sebastien Lahaie -
2013 Poster: Minimax Optimal Algorithms for Unconstrained Linear Optimization »
Brendan McMahan · Jacob D Abernethy -
2013 Poster: Adaptive Market Making via Online Learning »
Jacob D Abernethy · Satyen Kale -
2013 Poster: How to Hedge an Option Against an Adversary: Black-Scholes Pricing is Minimax Optimal »
Jacob D Abernethy · Peter Bartlett · Rafael Frongillo · Andre Wibisono -
2013 Spotlight: How to Hedge an Option Against an Adversary: Black-Scholes Pricing is Minimax Optimal »
Jacob D Abernethy · Peter Bartlett · Rafael Frongillo · Andre Wibisono -
2013 Oral: Adaptive Market Making via Online Learning »
Jacob D Abernethy · Satyen Kale -
2011 Poster: A Collaborative Mechanism for Crowdsourcing Prediction Problems »
Jacob D Abernethy · Rafael Frongillo -
2011 Oral: A Collaborative Mechanism for Crowdsourcing Prediction Problems »
Jacob D Abernethy · Rafael Frongillo -
2010 Poster: Repeated Games against Budgeted Adversaries »
Jacob D Abernethy · Manfred K. Warmuth