arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.09014cs.CV
收藏

自监督方法在跨域小样本学习中的表现如何?

How Well Do Self-Supervised Methods Perform in Cross-Domain Few-Shot Learning?

Artificial Intelligence Research Institute, Zhejiang Lab(之江实验室人工智能研究院);College of Computer and Information Engineering, Zhejiang Gongshang University(浙江工商大学计算机与信息工程学院)

Yiyi Zhang, Ying Zheng, Xiaogang Xu, Jun Wang

AI总结:

本文评估了自监督表示学习在跨域小样本学习(CDFSL)中的作用,发现自监督方法即使使用浅层架构或小数据集也能媲美SOTA方法,且比监督方法更具鲁棒性,并建议将Prototypical Classifier作为CDFSL的标准评估方案。

摘要:

跨域小样本学习(CDFSL)在计算机视觉领域仍是一个很大程度上未解决的问题,而自监督学习提供了一种有前景的解决方案。这两种学习方法都试图减轻深度网络对大规模标注数据的依赖。尽管自监督方法最近取得了巨大进展,但它们在CDFSL上的实用性相对未被探索。在本文中,我们通过对现有方法进行彻底评估,研究了自监督表示学习在CDFSL背景下的作用。令人惊讶的是,即使使用浅层架构或小型训练数据集,自监督方法也能与现有SOTA方法相比表现良好。然而,没有单一的自监督方法在所有数据集上都占主导地位,这表明现有的自监督方法并不普遍适用。此外,我们发现从自监督方法提取的表示比监督方法表现出更强的鲁棒性。有趣的是,自监督表示在源域上的表现与其在目标域上的适用性几乎没有相关性。作为我们研究的一部分,我们对六种代表性分类器的性能进行了客观测量。结果建议将Prototypical Classifier作为CDFSL的标准评估方案。

英文摘要:

Cross-domain few-shot learning (CDFSL) remains a largely unsolved problem in the area of computer vision, while self-supervised learning presents a promising solution. Both learning methods attempt to alleviate the dependency of deep networks on the requirement of large-scale labeled data. Although self-supervised methods have recently advanced dramatically, their utility on CDFSL is relatively unexplored. In this paper, we investigate the role of self-supervised representation learning in the context of CDFSL via a thorough evaluation of existing methods. It comes as a surprise that even with shallow architectures or small training datasets, self-supervised methods can perform favorably compared to the existing SOTA methods. Nevertheless, no single self-supervised approach dominates all datasets indicating that existing self-supervised methods are not universally applicable. In addition, we find that representations extracted from self-supervised methods exhibit stronger robustness than the supervised method. Intriguingly, whether self-supervised representations perform well on the source domain has little correlation with their applicability on the target domain. As part of our study, we conduct an objective measurement of the performance for six kinds of representative classifiers. The results suggest Prototypical Classifier as the standard evaluation recipe for CDFSL.

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。

↑