Towards Task Sampler Learning for Meta-Learning

TL;DR

This paper critically examines the role of task diversity in meta-learning, revealing that optimal performance depends on balancing diversity, entropy, and difficulty, and introduces an adaptive sampler to optimize task selection.

Contribution

It challenges the belief that more task diversity always improves meta-learning, and proposes a novel adaptive task sampler that dynamically balances task factors for better training.

Findings

No universal task sampling strategy guarantees optimal performance.

Over-constraining task diversity can cause under-fitting or over-fitting.

The generalization of meta-learning depends on task diversity, entropy, and difficulty.

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.