DATTA: Domain Diversity Aware Test-Time Adaptation for Dynamic Domain Shift Data Streams
It addresses a critical limitation in real-world deployment of machine learning models by enabling adaptation to changing data streams, though it is incremental as it builds on existing TTA frameworks.
The paper tackles the problem of test-time adaptation under dynamic domain shifts, where existing methods fail due to batch normalization errors and gradient conflicts, and proposes DATTA, which outperforms state-of-the-art methods by up to 13%.
Test-Time Adaptation (TTA) addresses domain shifts between training and testing. However, existing methods assume a homogeneous target domain (e.g., single domain) at any given time. They fail to handle the dynamic nature of real-world data, where single-domain and multiple-domain distributions change over time. We identify that performance drops in multiple-domain scenarios are caused by batch normalization errors and gradient conflicts, which hinder adaptation. To solve these challenges, we propose Domain Diversity Adaptive Test-Time Adaptation (DATTA), the first approach to handle TTA under dynamic domain shift data streams. It is guided by a novel domain-diversity score. DATTA has three key components: a domain-diversity discriminator to recognize single- and multiple-domain patterns, domain-diversity adaptive batch normalization to combine source and test-time statistics, and domain-diversity adaptive fine-tuning to resolve gradient conflicts. Extensive experiments show that DATTA significantly outperforms state-of-the-art methods by up to 13%. Code is available at https://github.com/DYW77/DATTA.