arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.03493cs.CVcs.LG
收藏

纹理驱动视觉学习中的低频捷径

Low-Frequency Shortcuts in Texture-Driven Visual Learning

Harvard University(哈佛大学);Kempner Institute(凯姆纳研究所)

Utku Şirin, Cathy Hou, Despina-Ekaterini Argiropoulos, David Alvarez-Melis, Stratos Idreos

AI总结:

本文分析了纹理驱动领域中神经网络依赖低频成分作为捷径的现象,提出通过裁剪低频成分来消除捷径,从而提升分布内准确率和鲁棒性。

摘要:

神经网络存在捷径学习问题,即学习到的特征在训练集上泛化良好,但在分布内(ID)或分布外(OOD)测试集上表现不佳。现有研究均基于少数几个标准基准,这些基准是形状驱动的。然而,许多应用领域是纹理驱动的。在这项工作中,我们针对纹理驱动领域进行了捷径学习分析,并将其与标准基准进行了比较。我们表明,纹理驱动领域存在低频捷径。它们主要基于少数具有偏斜频谱行为的低频成分(LFC)做出决策,尽管其分类信息存在于更高频率的细粒度细节中。从训练集和测试集中裁剪LFC可以消除捷径,并提供更平衡的频谱行为,将ID准确率提升高达8%。我们表明,低频捷径使模型极易受到OOD干扰的影响,导致与ID准确率相比下降高达70%。裁剪LFC显著提高了对低频干扰的鲁棒性,提升高达40%,并引入了对高频干扰的权衡;平衡的频谱行为提供了更好的泛化性能,而对高频特征的依赖增加则降低了泛化性能。OOD准确率取决于这两个因素之间的相互作用。

英文摘要:

Neural networks suffer from shortcut learning, where learned features generalize well to the training set but not to in-distribution (ID) or out-of-distribution (OOD) test sets. Existing studies are all based on a few standard benchmarks, which are shape-driven. Numerous application domains, however, are texture-driven. In this work, we present shortcut learning analysis for texture-driven domains and compare it with that of a standard benchmark. We show that texture-driven domains suffer from low-frequency shortcuts. They make the majority of their decisions based on a few low-frequency components (LFCs) with a skewed spectral behavior, despite that higher-frequency components (HFCs) have higher predictive power. Pruning LFCs from training and test sets mitigates the shortcut and provides a more balanced spectral behavior, improving the ID accuracy by up to 10% and OOD accuracy by up to 40% under algorithmic and real-world domain shifts. We show that general-purpose and domain-specific foundation models can also suffer from low-frequency shortcuts. While large models can mitigate the shortcuts, they incur a high computational cost and may result in a significantly lower accuracy than shortcut-pruned from-scratch trained small models. We show that reduced image resolutions amplify the degree of shortcuts; large frequency-transformation block sizes capture low-frequency shortcuts better than small block sizes; and, low-frequency shortcuts persist across different color spaces. Our findings provide valuable insights, which we hope will be useful for practitioners working on new, understudied domains.

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。

↑