Learn To Learn More Precisely

TL;DR

This paper introduces Meta Self-Distillation (MSD), a meta-learning framework that enhances models' ability to learn precise target knowledge and reduce noise influence, significantly improving few-shot classification accuracy.

Contribution

The paper proposes a novel meta-learning framework, MSD, that maximizes knowledge consistency to improve precision and generalization in few-shot learning tasks.

Findings

MSD improves few-shot classification accuracy.

MSD enhances knowledge consistency and robustness.

MSD performs well in standard and augmented scenarios.

Abstract

Meta-learning has been extensively applied in the domains of few-shot learning and fast adaptation, achieving remarkable performance. While Meta-learning methods like Model-Agnostic Meta-Learning (MAML) and its variants provide a good set of initial parameters for the model, the model still tends to learn shortcut features, which leads to poor generalization. In this paper, we propose the formal conception of "learn to learn more precisely", which aims to make the model learn precise target knowledge from data and reduce the effect of noisy knowledge, such as background and noise. To achieve this target, we proposed a simple and effective meta-learning framework named Meta Self-Distillation(MSD) to maximize the consistency of learned knowledge, enhancing the models' ability to learn precise target knowledge. In the inner loop, MSD uses different augmented views of the same support data to update the model respectively. Then in the outer loop, MSD utilizes the same query data to optimize the consistency of learned knowledge, enhancing the model's ability to learn more precisely. Our experiment demonstrates that MSD exhibits remarkable performance in few-shot classification tasks in both standard and augmented scenarios, effectively boosting the accuracy and consistency of knowledge learned by the model.