Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priors
We study the problem of generating adversarial examples in a black-box setting in which only loss-oracle access to a model is available. We introduce a framework that conceptually unifies much of the existing work on black-box attacks, and we demonstrate that the current state-of-the-art methods are optimal in a natural sense. Despite this optimality, we show how to improve black-box attacks by bringing a new element into the problem: gradient priors. We give a bandit optimization-based algorithm that allows us to seamlessly integrate any such priors, and we explicitly identify and incorporate two examples. The resulting methods use two to four times fewer queries and fail two to five times less often than the current state-of-the-art.
Code (3)
Similar Papers 제목 키워드 기반
Adversarial Attacks on Gaussian Process Bandits
Gaussian processes (GP) are a widely-adopted tool used to sequentially optimize black-box functions, where evaluations are costly and potentially noisy. Recent works on GP bandits have proposed to move beyond random nois…
Adversarial AttackGaussian ProcessesOpportunistic Target Selection: Early Directional Commitment for Query-Efficient Black-Box Adversarial Attacks
Black-box adversarial attacks that minimize only the ground-truth confidence suffer from class drift: perturbations wander through the feature space without committing to a specific adversarial class, wasting queries on …
Evaluation of Four Black-box Adversarial Attacks and Some Query-efficient Improvement Analysis
With the fast development of machine learning technologies, deep learning models have been deployed in almost every aspect of everyday life. However, the privacy and security of these models are threatened by adversarial…
Adversarial AttackEfficient Black-box Adversarial Attacks via Bayesian Optimization Guided by a Function Prior
This paper studies the challenging black-box adversarial attack that aims to generate adversarial examples against a black-box model by only using output feedback of the model to input queries. Some previous methods impr…
Adversarial AttackBayesian OptimizationEfficient Action Poisoning Attacks on Linear Contextual Bandits
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 algori…
Multi-Armed Bandits