对比式无源域自适应中被忽视的问题：在邻域上下文中利用源信息潜在增强
What Has Been Overlooked in Contrastive Source-Free Domain Adaptation: Leveraging Source-Informed Latent Augmentation within Neighborhood Context
AI总结：
针对无源域自适应中难以估计域差距的问题，本文提出一种基于源预训练模型引导的邻域潜在特征增强方法，通过提升正键信息量优化对比学习，在多个基准数据集上超越现有方法。
摘要：
无源域自适应（SFDA）涉及将最初使用有标签数据集（源域）训练的模型进行自适应调整，使其在无标签数据集（目标域）上有效运行，且在自适应过程中不依赖任何源数据。当两个域之间的数据分布存在显著差异，以及存在关于源模型训练数据的隐私担忧时，这种自适应尤为关键。在自适应过程中无法访问源数据，使得分析性地估计域差距变得具有挑战性。为了解决这个问题，已经提出了各种技术，例如无监督聚类、对比学习和持续学习。在本文中，我们首先对基于对比学习的SFDA进行了广泛的理论分析，主要是因为它已经展现出优于其他技术的性能。受所获见解的启发，我们随后引入了一种简单但高效的潜在增强方法，专为对比SFDA量身定制。这种增强方法在源预训练模型的引导下，利用查询样本邻域内潜在特征的分散性，以增强正键的信息量。我们的方法基于单一的基于InfoNCE的对比损失，在广泛认可的基准数据集上优于最先进的SFDA方法。
英文摘要：
Source-free domain adaptation (SFDA) involves adapting a model originally trained using a labeled dataset ({\em source domain}) to perform effectively on an unlabeled dataset ({\em target domain}) without relying on any source data during adaptation. This adaptation is especially crucial when significant disparities in data distributions exist between the two domains and when there are privacy concerns regarding the source model's training data. The absence of access to source data during adaptation makes it challenging to analytically estimate the domain gap. To tackle this issue, various techniques have been proposed, such as unsupervised clustering, contrastive learning, and continual learning. In this paper, we first conduct an extensive theoretical analysis of SFDA based on contrastive learning, primarily because it has demonstrated superior performance compared to other techniques. Motivated by the obtained insights, we then introduce a straightforward yet highly effective latent augmentation method tailored for contrastive SFDA. This augmentation method leverages the dispersion of latent features within the neighborhood of the query sample, guided by the source pre-trained model, to enhance the informativeness of positive keys. Our approach, based on a single InfoNCE-based contrastive loss, outperforms state-of-the-art SFDA methods on widely recognized benchmark 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。
