arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2404.08273cs.CVcs.CR
收藏

对抗防御难以推进?试试扩散模型

Struggle with Adversarial Defense? Try Diffusion

Yujie Li, Yanbin Wang, Haitao Xu, Bin Liu, Jianguo Sun, Zhenhao Guo, Wenrui Ma

AI总结:

针对现有扩散模型用于对抗防御存在的收敛难、成本高、易受自适应攻击等问题,提出基于贝叶斯定理的真值最大化扩散分类器及优化策略,在CIFAR10上强攻击下鲁棒准确率达SOTA。

摘要:

对抗攻击通过引入微小扰动引发分类错误。近年来,扩散模型被应用于图像分类器,通过对抗训练或净化对抗噪声来提升对抗鲁棒性。然而,基于扩散的对抗训练常面临收敛难题和计算成本高昂的问题。此外,基于扩散的净化方法不可避免地会导致数据偏移,且被认为易受更强的自适应攻击。为解决这些问题,我们提出了真值最大化扩散分类器(Truth Maximization Diffusion Classifier, TMDC),这是一种基于预训练扩散模型和贝叶斯定理的生成式贝叶斯分类器。与数据驱动的分类器不同,TMDC在贝叶斯原理的指导下,利用扩散模型的条件似然来确定输入图像的类别概率,从而免受数据偏移的影响和对抗训练的局限性。此外,为增强TMDC抵御更强对抗攻击的能力,我们提出了一种扩散分类器的优化策略。该策略包括在带真实标签作为条件的扰动数据集上对扩散模型进行后训练,引导扩散模型学习数据分布,并最大化真实标签下的似然。所提方法在CIFAR10数据集上针对强白盒攻击和强自适应攻击下取得了当前最优性能。具体而言,在ε=0.05时,TMDC针对l∞范数有界扰动的鲁棒准确率达到82.81%,针对l₂范数有界扰动的鲁棒准确率达到86.05%。

英文摘要:

Adversarial attacks induce misclassification by introducing subtle perturbations. Recently, diffusion models are applied to the image classifiers to improve adversarial robustness through adversarial training or by purifying adversarial noise. However, diffusion-based adversarial training often encounters convergence challenges and high computational expenses. Additionally, diffusion-based purification inevitably causes data shift and is deemed susceptible to stronger adaptive attacks. To tackle these issues, we propose the Truth Maximization Diffusion Classifier (TMDC), a generative Bayesian classifier that builds upon pre-trained diffusion models and the Bayesian theorem. Unlike data-driven classifiers, TMDC, guided by Bayesian principles, utilizes the conditional likelihood from diffusion models to determine the class probabilities of input images, thereby insulating against the influences of data shift and the limitations of adversarial training. Moreover, to enhance TMDC's resilience against more potent adversarial attacks, we propose an optimization strategy for diffusion classifiers. This strategy involves post-training the diffusion model on perturbed datasets with ground-truth labels as conditions, guiding the diffusion model to learn the data distribution and maximizing the likelihood under the ground-truth labels. The proposed method achieves state-of-the-art performance on the CIFAR10 dataset against heavy white-box attacks and strong adaptive attacks. Specifically, TMDC achieves robust accuracies of 82.81% against $l_{\infty}$ norm-bounded perturbations and 86.05% against $l_{2}$ norm-bounded perturbations, respectively, with $ε=0.05$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑