AI 中文总结
该研究针对表向图像分类器，评估了25种TTA技术在分布偏移下的效果，发现复合与光度类TTA可提升分布外性能，频域变换则会降低性能，为相关分类器鲁棒性提升提供了新方向。
AI 中文摘要
将表格数据转换为视觉表示的表向图像方法已成为利用深度学习模型高性能的新范式。尽管具有优势，但这些方法在分布偏移下的鲁棒性仍未得到充分探索。测试时增强（TTA）是图像分类中提升模型泛化性和鲁棒性的有效方法，即聚合每个输入的多个变换视图的预测。本研究评估了TTA技术对表向图像方法生成的表示在分布外（OOD）下预测性能的影响。考虑了六种表向图像编码方法：TINTO、IGTD、DeepInsight、BIE、DistanceMatrix、Fotomics。使用了25种TTA技术，分为六类：几何、光度、结构、频率/编码、Mixup、复合。采用了TableShift基准的两个数据集（HELOC和Voting），它们提供了分布内和OOD测试子集，旨在评估分布偏移对表格数据的影响。结果表明，TTA可提升OOD性能，其中复合和光度策略在鲁棒性与方差间提供最佳权衡；相比之下，改变编码器特征-强度映射的频域变换会持续降低性能。这些发现表明，TTA是提升基于表格数据生成的图像表示训练的分类器鲁棒性和泛化性的有前景方法，尤其在分布偏移场景下。
英文摘要
Tabular-to-image methods that convert tabular data into visual representations have emerged as a novel paradigm for leveraging the high performance of deep learning models. Despite their advantages, the robustness of these methods under distribution shifts remains under explored. Test-Time Augmentation (TTA) is an effective approach in image classification to improve model generalization and robustness, where predictions over multiple transformed views of each input are aggregated. This work evaluates the impact of TTA techniques on predictive performance under Out-Of-Distribution (OOD) for representations generated by tabular-to-image methods. Six tabular-to-image encoding methods were considered: TINTO, IGTD, DeepInsight, BIE, DistanceMatrix, Fotomics. Twenty-five TTA techniques were used, organized into six types: Geometric, Photometric, Structural, Frequency/Encoding, Mixup, and Composite. We employed two datasets from the TableShift benchmark (HELOC and Voting) that provide in-distribution and OOD test subsets designed to evaluate the effect of distribution shifts on tabular data. The results indicate that TTA improves OOD performance, with composite and photometric strategies providing the best trade-off between robustness and variance. In contrast, frequency-domain transformations that alter the encoder's feature-to-intensity mapping consistently degrade performance. These findings highlight TTA as a promising approach for improving the robustness and generalization of classifiers trained on image representations derived from tabular data, particularly under distribution shifts.
