Task Attended Meta-Learning for Few-Shot Learning

TL;DR

This paper introduces task attended meta-learning, a novel approach that weights tasks in batch training based on their importance, enhancing few-shot learning performance on complex datasets.

Contribution

It proposes a task attended meta-training method that dynamically weights tasks in batch learning, improving meta-model training efficiency and effectiveness.

Findings

Improved performance on miniImageNet and tieredImageNet datasets.

Task weighting enhances meta-learning outcomes.

The method is compatible with existing batch episodic training.

Abstract

Meta-learning (ML) has emerged as a promising direction in learning models under constrained resource settings like few-shot learning. The popular approaches for ML either learn a generalizable initial model or a generic parametric optimizer through episodic training. The former approaches leverage the knowledge from a batch of tasks to learn an optimal prior. In this work, we study the importance of a batch for ML. Specifically, we first incorporate a batch episodic training regimen to improve the learning of the generic parametric optimizer. We also hypothesize that the common assumption in batch episodic training that each task in a batch has an equal contribution to learning an optimal meta-model need not be true. We propose to weight the tasks in a batch according to their "importance" in improving the meta-model's learning. To this end, we introduce a training curriculum motivated by selective focus in humans, called task attended meta-training, to weight the tasks in a batch. Task attention is a standalone module that can be integrated with any batch episodic training regimen. The comparisons of the models with their non-task-attended counterparts on complex datasets like miniImageNet and tieredImageNet validate its effectiveness.