面向开放世界测试时自适应的可靠神经塌缩近似

中文摘要 AI 辅助

测试时自适应（Test-Time Adaptation, TTA）方法旨在弥合源域与目标域之间的域间隙。然而，传统TTA方法在标签分布偏移发生时会失效，这一挑战通常被称为开放世界场景。本文中，我们提出了一种名为可靠神经塌缩近似（Reliable Neural Collapse approximation, ReNC）的新方法用于开放世界测试时自适应（Open-World Test-Time Adaptation, OWTTA）。具体而言，我们利用神经塌缩（Neural Collapse, NC）作为结构先验以实现可靠的目标域自适应。在该先验的指导下，我们证明预训练分类器权重可作为源域的原型。通过测量样本与原型之间的相似度，我们过滤出分布外（Out-Of-Distribution, OOD）样本以进行可靠更新。此外，我们提出一种神经塌缩近似机制来优化这些原型，确保它们在逐渐适应目标域的同时保持神经塌缩结构。在多个开放世界基准上进行的大量实验证明了所提方法的优越性。我们的实证分析表明，ReNC能在目标域中更好地保留与NC相关的属性，为解释可靠OWTTA提供了有用证据，并为模型设计提供了新见解。代码可在指定URL获取。

英文摘要

Test-Time Adaptation (TTA) methods aim to bridge the domain gap between the source and target domains. However, traditional TTA methods become ineffective when the label distribution shift occurs, a challenge commonly referred to as an open-world scenario. In this paper, we introduce a new method named Reliable Neural Collapse approximation (ReNC) for Open-World Test-Time Adaptation (OWTTA). Specifically, we leverage neural collapse as a structural prior for reliable target-domain adaptation. Guided by this prior, we justify that the pre-trained classifier weights can serve as the prototypes of the source domain. By measuring the similarity between samples and prototypes, we filter out the Out-Of-Distribution~(OOD) samples for reliable updates. Furthermore, we propose a neural collapse approximation mechanism to refine these prototypes, ensuring they can gradually adapt to the target domain while maintaining the neural collapse structure. Extensive experiments on several open-world benchmarks demonstrate the superiority of the proposed method. Our empirical analysis suggests that ReNC better preserves NC-related properties in the target domain, providing useful evidence for explaining reliable OWTTA and offering new insights for model design. Code is available at https://github.com/JiaqiLin-AI/ReNC.

发表机构

• Centre for Frontier AI Research, Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局前沿人工智能研究中心）
• Sun Yat-sen University（中山大学）
• School of Computer Science and Engineering, Sun Yat-sen University（中山大学计算机科学与工程学院）
• School of Mathematics (Zhuhai), Sun Yat-sen University（中山大学（珠海校区）数学学院）

机构由 AI 辅助整理，请以论文原文为准。