arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2110.03374cs.CV
收藏

模型自适应:无源数据的无监督领域自适应历史对比学习

Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data

Nanyang Technological University(南洋理工大学)

Jiaxing Huang, Dayan Guan, Aoran Xiao, Shijian Lu

摘要:

无监督领域自适应的目的是对齐一个标记的源域和一个未标记的目标域,但它需要访问源数据,这往往会引起数据隐私,数据可移植性和数据传输效率的关注。我们研究了无监督模型自适应(UMA),或称为无源数据的无监督领域自适应,这是一种替代设置,旨在使源训练模型适应目标分布,而无需访问源数据。为此,我们设计了一个创新的历史对比学习(HCL)技术,利用历史源假设,以弥补缺乏源数据在UMA。HCL从两个角度应对UMA挑战。首先,它引入了历史对比实例判别(HCID),通过对比当前适应模型和历史模型生成的嵌入,从目标样本中学习。通过历史模型,HCID鼓励UMA学习实例判别目标表示,同时保留源假设。其次,它引入了历史对比类别歧视(HCCD),伪标签的目标样本学习类别歧视的目标表示。具体来说,HCCD根据伪标签在当前和历史模型中的预测一致性对其重新加权。大量的实验表明,HCL在各种视觉任务和设置中始终优于最先进的方法。

英文摘要:

Unsupervised domain adaptation aims to align a labeled source domain and an unlabeled target domain, but it requires to access the source data which often raises concerns in data privacy, data portability and data transmission efficiency. We study unsupervised model adaptation (UMA), or called Unsupervised Domain Adaptation without Source Data, an alternative setting that aims to adapt source-trained models towards target distributions without accessing source data. To this end, we design an innovative historical contrastive learning (HCL) technique that exploits historical source hypothesis to make up for the absence of source data in UMA. HCL addresses the UMA challenge from two perspectives. First, it introduces historical contrastive instance discrimination (HCID) that learns from target samples by contrasting their embeddings which are generated by the currently adapted model and the historical models. With the historical models, HCID encourages UMA to learn instance-discriminative target representations while preserving the source hypothesis. Second, it introduces historical contrastive category discrimination (HCCD) that pseudo-labels target samples to learn category-discriminative target representations. Specifically, HCCD re-weights pseudo labels according to their prediction consistency across the current and historical models. Extensive experiments show that HCL outperforms and state-of-the-art methods consistently across a variety of visual tasks and setups.

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。

↑