arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.04785cs.LGcs.AImath.CA
收藏

相位偏移对抗训练

Phase-shifted Adversarial Training

Deargen Inc.(Deargen公司);Korea Institute for Advanced Study(韩国高等研究院);Sungkyunkwan University(成均馆大学);Deargen USA Inc.(Deargen美国公司)

Yeachan Kim, Seongyeon Kim, Ihyeok Seo, Bonggun Shin

AI总结:

提出相位偏移对抗训练(PhaseAT),将高频成分移至低频范围以加速收敛,从而提升模型对抗鲁棒性,并在CIFAR-10和ImageNet上验证了有效性。

摘要:

对抗训练被认为是将基于神经网络的应用安全部署到现实世界中的必要组成部分。为了获得更强的鲁棒性,现有方法主要集中于通过增加更新步数、使用平滑损失函数对模型进行正则化以及向攻击中注入随机性来生成强攻击。相反,我们通过响应频率的视角分析对抗训练的行为。我们通过实验发现,对抗训练导致神经网络对高频信息的收敛性较低,从而在每个数据点附近产生高度振荡的预测。为了高效且有效地学习高频内容,我们首先证明了频率原理的普遍现象,即“低频先被学习”,在对抗训练中仍然成立。基于此,我们提出了相位偏移对抗训练(PhaseAT),该方法通过将这些频率偏移到快速收敛发生的低频范围,使模型学习高频成分。在评估方面,我们在CIFAR-10和ImageNet上进行了实验,并使用了为可靠评估而精心设计的自适应攻击。综合结果表明,PhaseAT显著提高了高频信息的收敛性。这通过使模型在每个数据点附近具有平滑的预测,从而提高了对抗鲁棒性。

英文摘要:

Adversarial training has been considered an imperative component for safely deploying neural network-based applications to the real world. To achieve stronger robustness, existing methods primarily focus on how to generate strong attacks by increasing the number of update steps, regularizing the models with the smoothed loss function, and injecting the randomness into the attack. Instead, we analyze the behavior of adversarial training through the lens of response frequency. We empirically discover that adversarial training causes neural networks to have low convergence to high-frequency information, resulting in highly oscillated predictions near each data. To learn high-frequency contents efficiently and effectively, we first prove that a universal phenomenon of frequency principle, i.e., \textit{lower frequencies are learned first}, still holds in adversarial training. Based on that, we propose phase-shifted adversarial training (PhaseAT) in which the model learns high-frequency components by shifting these frequencies to the low-frequency range where the fast convergence occurs. For evaluations, we conduct the experiments on CIFAR-10 and ImageNet with the adaptive attack carefully designed for reliable evaluation. Comprehensive results show that PhaseAT significantly improves the convergence for high-frequency information. This results in improved adversarial robustness by enabling the model to have smoothed predictions near each data.

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。

↑