Adversarial Eigen Attack on Black-Box Models
This work addresses a practical challenge in AI safety by enhancing black-box attack methods for scenarios with limited data access, though it is incremental as it builds on existing transferable attack frameworks.
The paper tackles the problem of improving black-box adversarial attack efficiency under a novel constraint where attackers can use a pre-trained white-box model's parameters but cannot retrain it with additional data, proposing the EigenBA algorithm that leverages the Jacobian matrix to find optimal perturbations, achieving significant efficiency gains on ImageNet and CIFAR-10 datasets.
Black-box adversarial attack has attracted a lot of research interests for its practical use in AI safety. Compared with the white-box attack, a black-box setting is more difficult for less available information related to the attacked model and the additional constraint on the query budget. A general way to improve the attack efficiency is to draw support from a pre-trained transferable white-box model. In this paper, we propose a novel setting of transferable black-box attack: attackers may use external information from a pre-trained model with available network parameters, however, different from previous studies, no additional training data is permitted to further change or tune the pre-trained model. To this end, we further propose a new algorithm, EigenBA to tackle this problem. Our method aims to explore more gradient information of the black-box model, and promote the attack efficiency, while keeping the perturbation to the original attacked image small, by leveraging the Jacobian matrix of the pre-trained white-box model. We show the optimal perturbations are closely related to the right singular vectors of the Jacobian matrix. Further experiments on ImageNet and CIFAR-10 show that even the unlearnable pre-trained white-box model could also significantly boost the efficiency of the black-box attack and our proposed method could further improve the attack efficiency.