用标准化框架重检无监督域适应关键假设，发现主流方法在先进模型上效果变差。
UDA-Bench: Revisiting Common Assumptions in Unsupervised Domain Adaptation Using a Standardized Framework
- 构建统一训练评估框架UDA-Bench，实现方法公平对比
- 先进骨干网络下自适应收益下降，未充分利用无标签数据
- 预训练数据对适配性能影响显著，适合领域迁移研究者
本文通过大规模、受控的实证研究，深入分析了影响现代无监督域适应（UDA）方法有效性的多种因素。为支持分析，我们首先开发了UDA-Bench——一个基于PyTorch的新框架，标准化了域适应的训练与评估流程，实现了多个UDA方法间的公平比较。利用UDA-Bench，我们的全面实证研究揭示：(i) 随着骨干网络的提升，自适应方法的优势减弱；(ii) 当前方法对无标签数据利用率不足；(iii) 预训练数据在监督与自监督设置下均显著影响下游适配性能。这些发现揭示了无监督适应中的若干新奇且出人意料的特性，同时科学验证了此前常被视为经验性直觉或实践技巧的若干假设。UDA-Bench框架及训练模型已公开于https://github.com/ViLab-UCSD/UDABench_ECCV2024。
原文摘要 · Abstract (English)
In this work, we take a deeper look into the diverse factors that influence the efficacy of modern unsupervised domain adaptation (UDA) methods using a large-scale, controlled empirical study. To facilitate our analysis, we first develop UDA-Bench, a novel PyTorch framework that standardizes training and evaluation for domain adaptation enabling fair comparisons across several UDA methods. Using UDA-Bench, our comprehensive empirical study into the impact of backbone architectures, unlabeled data quantity, and pre-training datasets reveals that: (i) the benefits of adaptation methods diminish with advanced backbones, (ii) current methods underutilize unlabeled data, and (iii) pre-training data significantly affects downstream adaptation in both supervised and self-supervised settings. In the context of unsupervised adaptation, these observations uncover several novel and surprising properties, while scientifically validating several others that were often considered empirical heuristics or practitioner intuitions in the absence of a standardized training and evaluation framework. The UDA-Bench framework and trained models are publicly available at https://github.com/ViLab-UCSD/UDABench_ECCV2024.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。