arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.04288cs.LG
收藏

并发分布偏移下的模型鲁棒性分析

An Analysis of Model Robustness across Concurrent Distribution Shifts

École Polytechnique Fédérale de Lausanne(洛桑联邦理工学院);Seoul National University(首尔大学);Samsung Research(三星研究院);Singapore-MIT Alliance for Research and Technology(新加坡-麻省理工学院研究与技术联盟)

Myeongho Jeon, Suhwan Choi, Hyoje Lee, Teresa Yeo

AI总结:

本文评估八种数据集168个源-目标对上的26种算法,分析多种并发分布偏移下的模型鲁棒性,发现启发式数据增强整体表现最佳。

摘要:

针对源数据精心优化的机器学习模型,在面临分布偏移(DSs)时往往无法准确预测目标数据。以往的基准研究虽然广泛,但主要集中于简单的分布偏移。鉴于现实场景中分布偏移通常以更复杂的形式出现,我们将研究范围扩展到多种并发偏移,例如未见域偏移与虚假相关性的组合。我们在来自八个数据集的168个源-目标对上评估了26种算法,涵盖从简单的启发式数据增强到利用基础模型进行零样本推理的方法。对超过10万个模型的分析表明:(i)与单一偏移相比,并发分布偏移通常会降低性能,但也存在某些例外;(ii)如果一个模型能够改善对某一种分布偏移的泛化,它往往也会对其他偏移有效;(iii)启发式数据增强在合成数据集和真实世界数据集上均取得了最佳的整体性能。

英文摘要:

Machine learning models, meticulously optimized for source data, often fail to predict target data when faced with distribution shifts (DSs). Previous benchmarking studies, though extensive, have mainly focused on simple DSs. Recognizing that DSs often occur in more complex forms in real-world scenarios, we broadened our study to include multiple concurrent shifts, such as unseen domain shifts combined with spurious correlations. We evaluated 26 algorithms that range from simple heuristic augmentations to zero-shot inference using foundation models, across 168 source-target pairs from eight datasets. Our analysis of over 100K models reveals that (i) concurrent DSs typically worsen performance compared to a single shift, with certain exceptions, (ii) if a model improves generalization for one distribution shift, it tends to be effective for others, and (iii) heuristic data augmentations achieve the best overall performance on both synthetic and real-world datasets.

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。

↑