Data Augmentation for Meta-Learning

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.

15 pages, 3 figures, Accepted to ICML 2021

References in corpus (7)

• Self-Augmentation: Generalizing Deep Networks to Unseen Classes for Few-Shot Learning
• Meta-Learning Requires Meta-Augmentation
• Assume, Augment and Learn: Unsupervised Few-Shot Meta-Learning via Random Labels and Data Augmentation
• MaxUp: A Simple Way to Improve Generalization of Neural Network Training
• Meta-Learned Confidence for Few-shot Learning
• Task Augmentation by Rotating for Meta-Learning
• Unraveling Meta-Learning: Understanding Feature Representations for Few-Shot Tasks

Cited by in corpus (5)

• Accelerating physics-informed neural network based 1D arc simulation by meta learning
• Meta-Learning with Fewer Tasks through Task Interpolation
• Semi-Supervised Few-Shot Intent Classification and Slot Filling
• Reducing and Exploiting Data Augmentation Noise through Meta Reweighting Contrastive Learning for Text Classification
• On Hard Episodes in Meta-Learning