基于双层极小极大优化的高效可靠迁移攻击学习
Learning with Bilevel-Minimax Optimization for Efficient and Reliable Transfer Attacks
浏览论文内容
中文总结 AI 辅助
该研究提出BMAT方法，通过双层极小极大优化实现初始化、代理适配与扰动的三元耦合，在30余种受害者模型上优于10余种基线，可高效可靠地生成迁移对抗样本。
中文摘要 AI 辅助
基于迁移的对抗攻击利用代理模型生成对抗样本，以误导黑盒受害者模型。除扰动生成外，可迁移性从根本上由初始化、代理适配与梯度动态的耦合关系决定。我们从双层极小极大视角重新审视这一挑战，提出BMAT（双层极小极大对抗迁移）方法。该双层公式捕捉初始化与扰动间的依赖关系，内层极小极大问题则提升代理模型的鲁棒性以实现跨架构泛化。算法层面，我们开发了一种集成式自底向上求解器，结合软权重调制器与隐式梯度近似器，实现初始化、代理适配与扰动优化三者的三元耦合。我们还为所提双层极小极大框架的优化动态提供了理论洞察。在分类与分割基准上开展的大量实验表明，BMAT在30余种受害者模型上的表现优于10余种强基线方法，同时提升了架构内与跨架构的可迁移性，并使mIoU降低最多达2倍。代码可在此https URL获取。
英文摘要
Transfer-based adversarial attacks craft adversarial examples using surrogate models to mislead black-box victim models. Beyond perturbation generation, transferability is fundamentally governed by the coupling of initialization, surrogate adaptation, and gradient dynamics. We revisit this challenge from a bilevel-minimax perspective and propose BMAT (Bilevel-Minimax Adversarial Transfer). The bilevel formulation captures the dependency between initialization and perturbation, while the inner minimax problem promotes surrogate robustness for cross-architecture generalization. Algorithmically, we develop an integrated bottom-up solver that combines a Soft Weight Modulator and an Implicit Gradient Approximator to enable ternary coupling among initialization, surrogate adaptation, and perturbation optimization. We further provide theoretical insights into the optimization dynamics of the proposed bilevel-minimax framework. Extensive experiments on classification and segmentation benchmarks show that BMAT outperforms more than 10 strong baselines across more than 30 victim models, improving both intra- and cross-architecture transfer and yielding up to a 2x reduction in mIoU. Code is available at https://github.com/callous-youth/BMAT.
发表机构
- The University of Hong Kong（香港大学）
- Dalian University of Technology（大连理工大学）
- The Hong Kong Polytechnic University（香港理工大学）
机构由 AI 辅助整理，请以论文原文为准。
