提出首个持续源域自适应方法，解决模型在多目标域中持续学习难题。
GMM-COMET: Continual Source-Free Universal Domain Adaptation via a Mean Teacher and Gaussian Mixture Model-Based Pseudo-Labeling
- 用均值教师框架结合高斯混合模型伪标签提升稳定性
- 在所有测试场景下持续优于仅用源数据训练的模型
- 适合需要长期适应新数据分布的工业级部署场景
无监督域适应旨在解决训练与测试数据间分布偏移导致神经网络性能下降的问题。在真实场景中，模型适应时可能无法访问源数据，且目标域标签空间可能与源域不同。这种设定称为源域自适应通用域适应（SF-UniDA），近期受到关注，但现有方法仅假设从源到目标存在单一域偏移。本文首次研究持续性SF-UniDA，即模型需连续适应一系列不同且未标注的目标域。基于先前在线SF-UniDA方法，我们结合其核心思想，在均值教师框架中引入基于高斯混合模型的伪标签策略以增强长期适应中的稳定性，并进一步设计一致性损失提升鲁棒性。所提方法GMM-COMET为持续性SF-UniDA提供了首个强基准，实验中唯一实现所有评估场景下持续优于仅使用源数据训练的模型。代码已开源：https://github.com/pascalschlachter/GMM-COMET。
原文摘要 · Abstract (English)
Unsupervised domain adaptation tackles the problem that domain shifts between training and test data impair the performance of neural networks in many real-world applications. Thereby, in realistic scenarios, the source data may no longer be available during adaptation, and the label space of the target domain may differ from the source label space. This setting, known as source-free universal domain adaptation (SF-UniDA), has recently gained attention, but all existing approaches only assume a single domain shift from source to target. In this work, we present the first study on continual SF-UniDA, where the model must adapt sequentially to a stream of multiple different unlabeled target domains. Building upon our previous methods for online SF-UniDA, we combine their key ideas by integrating Gaussian mixture model-based pseudo-labeling within a mean teacher framework for improved stability over long adaptation sequences. Additionally, we introduce consistency losses for further robustness. The resulting method GMM-COMET provides a strong first baseline for continual SF-UniDA and is the only approach in our experiments to consistently improve upon the source-only model across all evaluated scenarios. Our code is available at https://github.com/pascalschlachter/GMM-COMET.
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。