Learning Data Augmentation with Online Bilevel Optimization for Image Classification

TL;DR

This paper introduces an efficient bilevel optimization method to automatically learn optimal data augmentation strategies for image classification, reducing the need for manual tuning and external validation.

Contribution

It proposes a novel bilevel optimization framework that jointly learns data augmentation parameters and the classifier, improving accuracy without costly hyperparameter searches.

Findings

Achieves comparable or better accuracy than manual augmentation methods.

Eliminates the need for an external validation loop for hyperparameter tuning.

Demonstrates efficiency and effectiveness in image classification tasks.

Abstract

Data augmentation is a key practice in machine learning for improving generalization performance. However, finding the best data augmentation hyperparameters requires domain knowledge or a computationally demanding search. We address this issue by proposing an efficient approach to automatically train a network that learns an effective distribution of transformations to improve its generalization. Using bilevel optimization, we directly optimize the data augmentation parameters using a validation set. This framework can be used as a general solution to learn the optimal data augmentation jointly with an end task model like a classifier. Results show that our joint training method produces an image classification accuracy that is comparable to or better than carefully hand-crafted data augmentation. Yet, it does not need an expensive external validation loop on the data augmentation hyperparameters.