Use as Many Surrogates as You Want: Selective Ensemble Attack to Unleash Transferability without Sacrificing Resource Efficiency
Bo Yang, Hengwei Zhang, Jindong Wang, Yuchen Ren, Chenhao Lin, Chao Shen, Zhengyu Zhao

TL;DR
This paper introduces Selective Ensemble Attack (SEA), a method that dynamically selects diverse surrogate models across iterations to improve transferability of adversarial attacks without increasing resource consumption.
Contribution
The paper proposes a novel dynamic model selection strategy that decouples within-iteration and cross-iteration diversity, enhancing transferability efficiently.
Findings
SEA outperforms existing attacks in transferability by 8.5% on ImageNet.
SEA maintains resource efficiency by fixing within-iteration models.
The method generalizes well to real-world vision systems.
Abstract
In surrogate ensemble attacks, using more surrogate models yields higher transferability but lower resource efficiency. This practical trade-off between transferability and efficiency has largely limited existing attacks despite many pre-trained models are easily accessible online. In this paper, we argue that such a trade-off is caused by an unnecessary common assumption, i.e., all models should be \textit{identical} across iterations. By lifting this assumption, we can use as many surrogates as we want to unleash transferability without sacrificing efficiency. Concretely, we propose Selective Ensemble Attack (SEA), which dynamically selects diverse models (from easily accessible pre-trained models) across iterations based on our new interpretation of decoupling within-iteration and cross-iteration model diversity. In this way, the number of within-iteration models is fixed for…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdversarial Robustness in Machine Learning · Domain Adaptation and Few-Shot Learning · Explainable Artificial Intelligence (XAI)
