DeltaBound Attack: Efficient decision-based attack in low queries regime
This addresses the vulnerability of machine learning models to efficient attacks in real-world scenarios like remote APIs, though it is incremental in improving query efficiency.
The paper tackles the problem of adversarial attacks in the hard-label setting with limited queries, proposing the DeltaBound attack that achieves performance comparable to or better than state-of-the-art methods while using ≤1000 queries per example.
Deep neural networks and other machine learning systems, despite being extremely powerful and able to make predictions with high accuracy, are vulnerable to adversarial attacks. We proposed the DeltaBound attack: a novel, powerful attack in the hard-label setting with $\ell_2$ norm bounded perturbations. In this scenario, the attacker has only access to the top-1 predicted label of the model and can be therefore applied to real-world settings such as remote API. This is a complex problem since the attacker has very little information about the model. Consequently, most of the other techniques present in the literature require a massive amount of queries for attacking a single example. Oppositely, this work mainly focuses on the evaluation of attack's power in the low queries regime $\leq 1000$ queries) with $\ell_2$ norm in the hard-label settings. We find that the DeltaBound attack performs as well and sometimes better than current state-of-the-art attacks while remaining competitive across different kinds of models. Moreover, we evaluate our method against not only deep neural networks, but also non-deep learning models, such as Gradient Boosting Decision Trees and Multinomial Naive Bayes.