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 2409.20139cs.LGcs.CV
收藏

通过自然输入梯度刻画模型鲁棒性

Characterizing Model Robustness via Natural Input Gradients

MIT CSAIL(麻省理工学院计算机科学与人工智能实验室)

Adrián Rodríguez-Muñoz, Tongzhou Wang, Antonio Torralba

AI总结:

本文发现仅对自然样本的输入梯度范数进行正则化即可在视觉Transformer上高效提升鲁棒性,达到先进对抗训练90%以上性能且计算成本仅为其60%,并揭示了梯度集中到图像边缘也能显著增强鲁棒性。

摘要:

对抗鲁棒模型在每个数据样本附近是局部平滑的,因此微小扰动无法剧烈改变模型输出。在现代系统中,这种平滑性通常通过对抗训练获得,对抗训练明确强制模型在扰动样本上表现良好。在这项工作中,我们展示了仅对自然样本上的模型输入梯度进行正则化的惊人有效性。惩罚输入梯度范数通常被认为是一种远为逊色的方法。我们的分析表明,梯度范数正则化的性能关键取决于激活函数的平滑性,并且实际上在采用平滑激活而非分段线性激活(例如 ReLU)的现代视觉Transformer上极其有效,这与先前的看法相反。在 ImageNet-1k 上,梯度范数训练达到了最先进的 PGD-3 对抗训练性能的 90% 以上(52% 对比约 56%),同时仅使用最先进方法 60% 的计算成本,且无需复杂的对抗优化。我们的分析还揭示了模型鲁棒性与自然输入梯度性质之间的关系,例如不对称的样本统计和通道统计。令人惊讶的是,我们发现仅通过正则化梯度使其集中在图像边缘上,而无需显式约束梯度范数,就能显著提高模型鲁棒性。

英文摘要:

Adversarially robust models are locally smooth around each data sample so that small perturbations cannot drastically change model outputs. In modern systems, such smoothness is usually obtained via Adversarial Training, which explicitly enforces models to perform well on perturbed examples. In this work, we show the surprising effectiveness of instead regularizing the gradient with respect to model inputs on natural examples only. Penalizing input Gradient Norm is commonly believed to be a much inferior approach. Our analyses identify that the performance of Gradient Norm regularization critically depends on the smoothness of activation functions, and are in fact extremely effective on modern vision transformers that adopt smooth activations over piecewise linear ones (eg, ReLU), contrary to prior belief. On ImageNet-1k, Gradient Norm training achieves > 90% the performance of state-of-the-art PGD-3 Adversarial Training} (52% vs.~56%), while using only 60% computation cost of the state-of-the-art without complex adversarial optimization. Our analyses also highlight the relationship between model robustness and properties of natural input gradients, such as asymmetric sample and channel statistics. Surprisingly, we find model robustness can be significantly improved by simply regularizing its gradients to concentrate on image edges without explicit conditioning on the gradient norm.

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。

↑