论文标题
频率偏见如何影响神经图像分类器的鲁棒性，以防止常见的腐败和对抗性扰动？
How Does Frequency Bias Affect the Robustness of Neural Image Classifiers against Common Corruption and Adversarial Perturbations?
论文作者
论文摘要
模型鲁棒性对于在现实世界应用中可靠的机器学习模型的可靠部署至关重要。最近的研究表明，数据增强可以导致模型过度依赖低频域中的特征，从而牺牲了低频腐败的性能，从而突出了频率与稳健性之间的联系。在这里，我们迈出了进一步的一步，可以通过其雅各布人的镜头更直接地研究模型的频率偏差及其对模型鲁棒性的影响。为了实现这一目标，我们建议模型的Jacobians的Jacobian频率正规化具有较大的低频组件比例。通过在四个图像数据集上的实验，我们表明分类器偏向低（高） - 频率组件可以为高（低） - 频率腐败和对抗性扰动带来性能增益，尽管在低（高） - 频率腐败方面取决于绩效。我们的方法阐明了深度学习模型的频率偏差和鲁棒性之间的更直接联系。
Model robustness is vital for the reliable deployment of machine learning models in real-world applications. Recent studies have shown that data augmentation can result in model over-relying on features in the low-frequency domain, sacrificing performance against low-frequency corruptions, highlighting a connection between frequency and robustness. Here, we take one step further to more directly study the frequency bias of a model through the lens of its Jacobians and its implication to model robustness. To achieve this, we propose Jacobian frequency regularization for models' Jacobians to have a larger ratio of low-frequency components. Through experiments on four image datasets, we show that biasing classifiers towards low (high)-frequency components can bring performance gain against high (low)-frequency corruption and adversarial perturbation, albeit with a tradeoff in performance for low (high)-frequency corruption. Our approach elucidates a more direct connection between the frequency bias and robustness of deep learning models.
