Optimizing Data Augmentation Policy Through Random Unidimensional Search

2021/06/16 by Xiaomeng Dong, Dong, Xiaomeng, Michael Potter +8
Computer Science · #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Privacy-Preserving Technologies in Data

paper · pdf · doi:10.48550/arxiv.2106.08756

openalex publication_date 2021/06/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

It is no secret amongst deep learning researchers that finding the optimal data augmentation strategy during training can mean the difference between state-of-the-art performance and a run-of-the-mill result. To that end, the community has seen many efforts to automate the process of finding the perfect augmentation procedure for any task at hand. Unfortunately, even recent cutting-edge methods bring massive computational overhead, requiring as many as 100 full model trainings to settle on an ideal configuration. We show how to achieve equivalent performance using just 6 trainings with Random Unidimensional Augmentation. Source code is available at https://github.com/fastestimator/RUA/tree/v1.0

Citations

• Wide Residual Networks
• Improved Regularization of Convolutional Neural Networks with Cutout
• YOLOv4: Optimal Speed and Accuracy of Object Detection
• Learning Multiple Layers of Features from Tiny Images

Related

• Learning data augmentation policies using augmented random search
• AugMax: Adversarial Composition of Random Augmentations for Robust\n Training
• When Chosen Wisely, More Data Is What You Need: A Universal Sample-Efficient Strategy For Data Augmentation
• RandAugment: Practical automated data augmentation with a reduced search space
• DADA: Differentiable Automatic Data Augmentation