arXiv 2502.206772025-02-28cs.LG

FoCTTA：聚焦的低内存持续测试时自适应

FoCTTA: Low-Memory Continual Test-Time Adaptation with Focus

Faculty of Computing, Harbin Institute of Technology（哈尔滨工业大学计算机学院）；Swiss Data Science Center（瑞士数据科学中心）；City University of Hong Kong（香港城市大学）

Youbing Hu, Yun Cheng, Zimu Zhou, Anqi Lu, Zhiqiang Cao, Zhijun Li

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对持续测试时自适应中更新所有BN层导致的高内存问题，提出FoCTTA，通过自动聚焦更新少数漂移敏感表征层，降低内存并提升准确率，在多个基准上优于现有方法。

摘要：

持续适应测试时的域偏移（CTTA）对于增强深度学习驱动的物联网应用的智能性至关重要。然而，现有的TTA方法通常更新所有批归一化（BN）层，表现出两种内存低效。首先，依赖BN层进行适应需要较大的批次大小，导致高内存使用。其次，更新所有BN层需要存储所有BN层的激活值以进行反向传播，加剧了内存需求。这两个因素导致显著的内存成本，使得现有解决方案在物联网设备上不切实际。在本文中，我们提出了FoCTTA，一种低内存的CTTA策略。关键在于自动识别并适应少数漂移敏感的表征层，而不是盲目更新所有BN层。从BN层转向表征层消除了对大批次大小的需求。此外，通过仅更新适应关键层，FoCTTA避免了存储过多的激活值。这种聚焦的适应方法确保FoCTTA不仅内存高效，而且保持有效的适应。评估显示，在相同内存约束下，FoCTTA在CIFAR10-C、CIFAR100-C和ImageNet-C上的适应准确率分别比现有最先进方法提高了4.5%、4.9%和14.8%。在各种批次大小下，FoCTTA平均将内存使用减少了3倍，同时在三个数据集上的准确率分别提高了8.1%、3.6%和0.2%。

英文摘要：

Continual adaptation to domain shifts at test time (CTTA) is crucial for enhancing the intelligence of deep learning enabled IoT applications. However, prevailing TTA methods, which typically update all batch normalization (BN) layers, exhibit two memory inefficiencies. First, the reliance on BN layers for adaptation necessitates large batch sizes, leading to high memory usage. Second, updating all BN layers requires storing the activations of all BN layers for backpropagation, exacerbating the memory demand. Both factors lead to substantial memory costs, making existing solutions impractical for IoT devices. In this paper, we present FoCTTA, a low-memory CTTA strategy. The key is to automatically identify and adapt a few drift-sensitive representation layers, rather than blindly update all BN layers. The shift from BN to representation layers eliminates the need for large batch sizes. Also, by updating adaptation-critical layers only, FoCTTA avoids storing excessive activations. This focused adaptation approach ensures that FoCTTA is not only memory-efficient but also maintains effective adaptation. Evaluations show that FoCTTA improves the adaptation accuracy over the state-of-the-arts by 4.5%, 4.9%, and 14.8% on CIFAR10-C, CIFAR100-C, and ImageNet-C under the same memory constraints. Across various batch sizes, FoCTTA reduces the memory usage by 3-fold on average, while improving the accuracy by 8.1%, 3.6%, and 0.2%, respectively, on the three 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。