arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.07899cs.CVcs.LG
收藏

无标记数据的跨域少样本学习

Cross-domain few-shot learning with unlabelled data

iFlyTek-Surrey Joint Research Centre on Artificial Intelligence(科大讯飞-萨里人工智能联合研究中心)

Fupin Yao

摘要:

少样本学习旨在解决数据稀缺问题。如果测试集和训练集之间存在域偏移,则它们的性能将大大降低。此设置称为跨域少样本学习。然而,这是非常具有挑战性的,因为目标域在训练期间是不可见的。因此,我们提出了一种新的设置一些未标记的数据从目标域提供,它可以弥合源域和目标域之间的差距。使用DomainNet \cite{peng2018oment}构建此设置的基准。我们提出了一种自监督学习方法,充分利用标记训练集和未标记训练集中的知识。大量的实验表明,我们的方法优于几个基线方法的大幅度。我们还精心设计了一个情景训练管道,可以显著提高性能。

英文摘要:

Few shot learning aims to solve the data scarcity problem. If there is a domain shift between the test set and the training set, their performance will decrease a lot. This setting is called Cross-domain few-shot learning. However, this is very challenging because the target domain is unseen during training. Thus we propose a new setting some unlabelled data from the target domain is provided, which can bridge the gap between the source domain and the target domain. A benchmark for this setting is constructed using DomainNet \cite{peng2018oment}. We come up with a self-supervised learning method to fully utilize the knowledge in the labeled training set and the unlabelled set. Extensive experiments show that our methods outperforms several baseline methods by a large margin. We also carefully design an episodic training pipeline which yields a significant performance boost.

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。

↑