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 2410.04201cs.CV
收藏

IT$^3$:幂等测试时训练

IT$^3$: Idempotent Test-Time Training

EPFL(瑞士洛桑联邦理工学院);NVIDIA(英伟达);Bilkent University(比尔肯特大学);UC Berkeley(加州大学伯克利分校)

Nikita Durasov, Assaf Shocher, Doruk Oner, Gal Chechik, Alexei A. Efros, Pascal Fua

AI总结:

本文提出幂等测试时训练(IT$^3$),利用幂等性替代辅助任务,仅用当前测试实例实现分布偏移的实时适应,并在多个领域和架构上优于现有方法。

摘要:

深度学习模型在部署到现实世界环境时,常常因训练数据和测试数据之间的分布偏移而表现不佳。虽然现有的方法如领域适应和测试时训练(TTT)提供了部分解决方案,但它们通常需要额外的数据或特定领域的辅助任务。我们提出了幂等测试时训练(IT$^3$),一种新颖的方法,仅使用当前测试实例即可实时适应分布偏移,无需任何辅助任务设计。我们的关键见解是,强制幂等性——即重复应用函数产生相同结果——可以有效替代先前TTT方法中使用的领域特定辅助任务。我们在理论上将幂等性与预测置信度联系起来,并证明在推理过程中最小化模型连续应用之间的距离可以改善分布外性能。在多个领域(包括图像分类、气动预测和航空分割)和架构(MLPs、CNNs、GNNs)上的大量实验表明,IT$^3$在更简单且更广泛适用的情况下,始终优于现有方法。我们的结果表明,幂等性为测试时适应提供了一个通用原则,可跨领域和架构进行泛化。

英文摘要:

Deep learning models often struggle when deployed in real-world settings due to distribution shifts between training and test data. While existing approaches like domain adaptation and test-time training (TTT) offer partial solutions, they typically require additional data or domain-specific auxiliary tasks. We present Idempotent Test-Time Training (IT$^3$), a novel approach that enables on-the-fly adaptation to distribution shifts using only the current test instance, without any auxiliary task design. Our key insight is that enforcing idempotence -- where repeated applications of a function yield the same result -- can effectively replace domain-specific auxiliary tasks used in previous TTT methods. We theoretically connect idempotence to prediction confidence and demonstrate that minimizing the distance between successive applications of our model during inference leads to improved out-of-distribution performance. Extensive experiments across diverse domains (including image classification, aerodynamics prediction, and aerial segmentation) and architectures (MLPs, CNNs, GNNs) show that IT$^3$ consistently outperforms existing approaches while being simpler and more widely applicable. Our results suggest that idempotence provides a universal principle for test-time adaptation that generalizes across domains and architectures.

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。

↑