Query-Efficient Black-box Adversarial Examples (superceded)
This addresses the vulnerability of neural network classifiers to practical black-box attacks, with incremental improvements in efficiency and applicability.
The paper tackles the problem of generating adversarial examples in black-box settings with limited queries and partial information, achieving a reduction of two to three orders of magnitude in query requirements and successfully performing the first targeted attack on the Google Cloud Vision API.
Note that this paper is superceded by "Black-Box Adversarial Attacks with Limited Queries and Information." Current neural network-based image classifiers are susceptible to adversarial examples, even in the black-box setting, where the attacker is limited to query access without access to gradients. Previous methods --- substitute networks and coordinate-based finite-difference methods --- are either unreliable or query-inefficient, making these methods impractical for certain problems. We introduce a new method for reliably generating adversarial examples under more restricted, practical black-box threat models. First, we apply natural evolution strategies to perform black-box attacks using two to three orders of magnitude fewer queries than previous methods. Second, we introduce a new algorithm to perform targeted adversarial attacks in the partial-information setting, where the attacker only has access to a limited number of target classes. Using these techniques, we successfully perform the first targeted adversarial attack against a commercially deployed machine learning system, the Google Cloud Vision API, in the partial information setting.