Adaptive few-shot learning with a fair priori distribution. (September 2022)

• Record Type: Journal Article
• Authors: Zeng, Xinke; Huang, Bo; Jia, Ke; Jia, Li; Zhao, Ke
• Is Part Of: Computers & electrical engineering. Volume 102(2022)
• Publication Date: 2022-09
• Subjects: Few-shot learning -- Adaptive learning -- A priori knowledge -- Metric learning -- Hard example mining; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108133
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Few-shot learning is a special challenge in pattern recognition, which identifies unseen categories given only limited samples. In the past few years, various methods have been proposed to solve few-shot problems. Based on the predecessors, we discuss the prior influence of the training set bring into the model, the relationship between data quality and feature performance, as well as the method and reason to expand negative samples. We consequently improve the models and learning strategies in these aspects, then we obtain a novel framework named AFSL: Adaptive Few-Shot Learning, which can eliminate the impact of data sets on the model, and also performs well on several public datasets. Specifically, we achieved a percentage increase of verification accuracy of 7.8% and 1.5% for 1-shot and 5-shots tasks comparing to prototype network on mini-Imagenet dataset, 9.8% and 2.4% on tiered-Imagenet, and 28.4% and 14.0% on the CUB dataset. Highlights: Training data will bring serious bias information to classification results. The image quality of sample is positively correlated with its feature's normalization. An adaptive approach should be used to balance the effects of samples. More negative samples can help to compress the within class distance, to enhance the discriminative capability of the model.