Improving Few-Shot Learning through Multi-task Representation Learning Theory

TL;DR

This paper leverages multi-task representation learning theory to enhance few-shot classification, introducing a spectral regularization technique inspired by theoretical insights and validating it through experiments.

Contribution

It applies recent MTR theory to practical meta-learning, proposing a spectral regularization to improve few-shot learning performance.

Findings

Spectral regularization improves meta-learning accuracy

Theoretical analysis explains differences between gradient and metric-based methods

Experimental results confirm the effectiveness of the proposed method

Abstract

In this paper, we consider the framework of multi-task representation (MTR) learning where the goal is to use source tasks to learn a representation that reduces the sample complexity of solving a target task. We start by reviewing recent advances in MTR theory and show that they can provide novel insights for popular meta-learning algorithms when analyzed within this framework. In particular, we highlight a fundamental difference between gradient-based and metric-based algorithms in practice and put forward a theoretical analysis to explain it. Finally, we use the derived insights to improve the performance of meta-learning methods via a new spectral-based regularization term and confirm its efficiency through experimental studies on few-shot classification benchmarks. To the best of our knowledge, this is the first contribution that puts the most recent learning bounds of MTR theory into practice for the task of few-shot classification.