HopSkipJumpAttack: A Query-Efficient Decision-Based Attack
By Jianbo Chen, Michael I. Jordan. 2019
835 citations recorded by the source corpus.
The goal of a decision-based adversarial attack on a trained model is to generate adversarial examples based solely on observing output labels returned by the targeted model. We develop HopSkipJumpAttack, a family of algorithms based on a novel estimate of the gradient direction using binary information at the decision boundary. The proposed family includes both untargeted and targeted attacks optimized for ℓ and ℓ∞ similarity metrics respectively. Theoretical analysis is provided for the proposed algorithms and the gradient direction estimate. Experiments show HopSkipJumpAttack requires significantly fewer model queries than several state-of-the-art decision-based adversarial attacks. It also achieves competitive performance in attacking several widely-used defense mechanisms.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Ajmal Saeed Mian
- Florian Tramèr
- Nicholas Carlini
- Nicolas Papernot
- Naveed Akhtar
- Baoyuan Wu
- Mubarak Shah
- Christopher A. Choquette-Choo
Citations in the corpus
- Label-Only Membership Inference Attacks (2020, methodology): “Remarkably, we find that our label-only boundary distance attack—based on the HopSkipJump attack [44] for finding adversarial examples—performs on-par with, or slightly better than, the confidence-vector attacks, despite having access to a more restricted query interface.”
- Advances in adversarial attacks and defenses in computer vision: A survey (2021, result): “Better performance in terms of the number of queries and perturbation norm are reported as compared to the Boundary attack [74], HopSkipJump attack [75] and the qFool attack [76].”
- Boosting Decision-Based Black-Box Adversarial Attacks with Random Sign Flip (2020, methodology): “For other defense mechanisms, we only compare with the existing state-ofthe-art method HSJA [10].”
- Adversarial Machine Learning for Network Intrusion Detection Systems: A Comprehensive Survey (2023, background): “tive against the C & W attack [91] and the Hop-Skip-Jump attack [95].”