Unsupervised meta-learning for few-shot learning

doi:10.1016/j.patcog.2021.107951

	Unsupervised meta-learning for few-shot learning
	Xu, Hui 1; Wang, Jiaxing2 ; Li, Hao 1; Ouyang, Deqiang 1; Shao, Jie 1,3
发表期刊	PATTERN RECOGNITION
ISSN	0031-3203
	2021-08-01
卷号	116 页码:10
通讯作者	Shao, Jie(shaojie@uestc.edu.cn)
摘要	Meta-learning is an effective tool to address the few-shot learning problem, which requires new data to be classified considering only a few training examples. However, when used for classification, it requires large labeled datasets, which are not always available in practice. In this paper, we propose an unsupervised meta-learning algorithm that learns from an unlabeled dataset and adapts to downstream human specific tasks with few labeled data. The proposed algorithm constructs tasks using clustering embedding methods and data augmentation functions to satisfy two critical class distinction requirements. To alleviate the biases and the weak diversity problem introduced by data augmentation functions, the proposed algorithm uses two methods, which are shifting the feeding data between the inner-outer loops and a novel data augmentation function. We further provide theoretical analysis of the effect of augmentation data in the inner/outer loop. Experiments on the MiniImagenet and Omniglot datasets demonstrate that the proposed unsupervised meta-learning approach outperforms other tested unsupervised representation learning approaches and two recent unsupervised meta-learning baselines. Compared with supervised meta-learning approaches, certain results produced by our method are quite close to those produced by such methods trained on the human-designed labeled tasks. (c) 2021 Elsevier Ltd. All rights reserved.
关键词	Unsupervised learning Meta-learning Few-shot learning
DOI	10.1016/j.patcog.2021.107951
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2018AAA0102200] ; National Natural Science Foundation of China[61832001] ; Sichuan Science and Technology Program[2019YFG0535] ; Sichuan Science and Technology Program[2021JDRC0079]
项目资助者	National Key Research and Development Program of China ; National Natural Science Foundation of China ; Sichuan Science and Technology Program
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000655616200003
出版者	ELSEVIER SCI LTD
引用统计	被引频次：33[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/45299
专题	复杂系统认知与决策实验室_先进机器人
通讯作者	Shao, Jie
作者单位	1.Univ Elect Sci & Technol China, Sch Comp Sci & Engn, Ctr Future Media, Chengdu 611731, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 3.Sichuan Artificial Intelligence Res Inst, Yibin 644000, Peoples R China
推荐引用方式 GB/T 7714	Xu, Hui,Wang, Jiaxing,Li, Hao,et al. Unsupervised meta-learning for few-shot learning[J]. PATTERN RECOGNITION,2021,116:10.
APA	Xu, Hui,Wang, Jiaxing,Li, Hao,Ouyang, Deqiang,&Shao, Jie.(2021).Unsupervised meta-learning for few-shot learning.PATTERN RECOGNITION,116,10.
MLA	Xu, Hui,et al."Unsupervised meta-learning for few-shot learning".PATTERN RECOGNITION 116(2021):10.