Deep AutoAugment

TL;DR

Deep AutoAugment introduces a fully automated, multi-layer data augmentation pipeline that optimizes augmentation policies from scratch using gradient similarity, achieving competitive results without relying on predefined augmentations.

Contribution

It presents a novel automated approach for data augmentation search that builds a multi-layer pipeline from scratch, removing the need for hand-picked default augmentations.

Findings

Learned augmentation policies achieve strong performance comparable to previous methods.

Regularized gradient matching effectively guides augmentation policy search.

The method eliminates reliance on predefined default augmentations.

Abstract

While recent automated data augmentation methods lead to state-of-the-art results, their design spaces and the derived data augmentation strategies still incorporate strong human priors. In this work, instead of fixing a set of hand-picked default augmentations alongside the searched data augmentations, we propose a fully automated approach for data augmentation search named Deep AutoAugment (DeepAA). DeepAA progressively builds a multi-layer data augmentation pipeline from scratch by stacking augmentation layers one at a time until reaching convergence. For each augmentation layer, the policy is optimized to maximize the cosine similarity between the gradients of the original and augmented data along the direction with low variance. Our experiments show that even without default augmentations, we can learn an augmentation policy that achieves strong performance with that of previous works. Extensive ablation studies show that the regularized gradient matching is an effective search method for data augmentation policies. Our code is available at: https://github.com/MSU-MLSys-Lab/DeepAA .