Back to Search Start Over

Towards Task Sampler Learning for Meta-Learning.

Authors :
Wang, Jingyao
Qiang, Wenwen
Su, Xingzhe
Zheng, Changwen
Sun, Fuchun
Xiong, Hui
Source :
International Journal of Computer Vision. Dec2024, Vol. 132 Issue 12, p5534-5564. 31p.
Publication Year :
2024

Abstract

Meta-learning aims to learn general knowledge with diverse training tasks conducted from limited data, and then transfer it to new tasks. It is commonly believed that increasing task diversity will enhance the generalization ability of meta-learning models. However, this paper challenges this view through empirical and theoretical analysis. We obtain three conclusions: (i) there is no universal task sampling strategy that can guarantee the optimal performance of meta-learning models; (ii) over-constraining task diversity may incur the risk of under-fitting or over-fitting during training; and (iii) the generalization performance of meta-learning models are affected by task diversity, task entropy, and task difficulty. Based on this insight, we design a novel task sampler, called Adaptive Sampler (ASr). ASr is a plug-and-play module that can be integrated into any meta-learning framework. It dynamically adjusts task weights according to task diversity, task entropy, and task difficulty, thereby obtaining the optimal probability distribution for meta-training tasks. Finally, we conduct experiments on a series of benchmark datasets across various scenarios, and the results demonstrate that ASr has clear advantages. The code is publicly available at https://github.com/WangJingyao07/Adaptive-Sampler. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
09205691
Volume :
132
Issue :
12
Database :
Academic Search Index
Journal :
International Journal of Computer Vision
Publication Type :
Academic Journal
Accession number :
180936120
Full Text :
https://doi.org/10.1007/s11263-024-02145-0