Data Augmentation for Meta-Learning

TL;DR

This paper explores how data augmentation can be effectively integrated into meta-learning pipelines at both image and class levels, significantly enhancing few-shot classification performance.

Contribution

It systematically analyzes data augmentation strategies in meta-learning and proposes a meta-specific augmentation method that improves few-shot learning results.

Findings

Meta-specific data augmentation improves meta-learner performance

Augmentation at class level generates new tasks and classes

Enhanced accuracy on few-shot classification benchmarks

Abstract

Conventional image classifiers are trained by randomly sampling mini-batches of images. To achieve state-of-the-art performance, practitioners use sophisticated data augmentation schemes to expand the amount of training data available for sampling. In contrast, meta-learning algorithms sample support data, query data, and tasks on each training step. In this complex sampling scenario, data augmentation can be used not only to expand the number of images available per class, but also to generate entirely new classes/tasks. We systematically dissect the meta-learning pipeline and investigate the distinct ways in which data augmentation can be integrated at both the image and class levels. Our proposed meta-specific data augmentation significantly improves the performance of meta-learners on few-shot classification benchmarks.