Amortized Bayesian Prototype Meta-learning: A New Probabilistic Meta-learning Approach to Few-shot Image Classification

CORC > 兰州理工大学 > 兰州理工大学

	Amortized Bayesian Prototype Meta-learning: A New Probabilistic Meta-learning Approach to Few-shot Image Classification
	Sun, Zhuo 1; Wu, Jijie 2; Li, Xiaoxu 2; Yang, Wenming 3; Xue, Jing-Hao 1
	2021
会议日期	APR 13-15, 2021
会议地点	ELECTR NETWORK
卷号	130
英文摘要	Probabilistic meta-learning methods recently have achieved impressive success in few-shot image classification. However, they introduce a huge number of random variables for neural network weights and thus severe computational and inferential challenges. In this paper, we propose a novel probabilistic meta-learning method called amortized Bayesian prototype meta-learning. In contrast to previous methods, we introduce only a small number of random variables for latent class prototypes rather than a huge number for network weights; we learn to learn the posterior distributions of these latent prototypes in an amortized inference way with no need for an extra amortization network, such that we can easily approximate their posteriors conditional on few labeled samples, whenever at meta-training or meta-testing stage. The proposed method can be trained end-to-end without any pre-training. Compared with other probabilistic meta-learning methods, our proposed approach is more interpretable with much less random variables, while still be able to achieve competitive performance for few-shot image classification problems on various benchmark datasets. Its excellent robustness and predictive uncertainty are also demonstrated through ablation studies.
会议录	24TH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS (AISTATS)
会议录出版者	MICROTOME PUBLISHING
会议录出版地	BROOKLINE
语种	英语
ISSN号	2640-3498
WOS研究方向	Computer Science ; Mathematics
WOS记录号	WOS:000659893801071
内容类型	会议论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/150129]
专题	兰州理工大学
作者单位	1.UCL, London, England; 2.Lanzhou Univ Technol, Lanzhou, Peoples R China; 3.Tsinghua Univ, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Sun, Zhuo,Wu, Jijie,Li, Xiaoxu,et al. Amortized Bayesian Prototype Meta-learning: A New Probabilistic Meta-learning Approach to Few-shot Image Classification[C]. 见:. ELECTR NETWORK. APR 13-15, 2021.