Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
This addresses the computational infeasibility of state-of-the-art augmentation methods for ordinary users, offering a more efficient alternative.
The paper tackles the challenge of efficiently selecting data augmentation policies for neural network training by introducing Population Based Augmentation (PBA), which generates nonstationary schedules and matches AutoAugment's performance on datasets like CIFAR-10 with a 1.46% test error while using three orders of magnitude less compute.
A key challenge in leveraging data augmentation for neural network training is choosing an effective augmentation policy from a large search space of candidate operations. Properly chosen augmentation policies can lead to significant generalization improvements; however, state-of-the-art approaches such as AutoAugment are computationally infeasible to run for the ordinary user. In this paper, we introduce a new data augmentation algorithm, Population Based Augmentation (PBA), which generates nonstationary augmentation policy schedules instead of a fixed augmentation policy. We show that PBA can match the performance of AutoAugment on CIFAR-10, CIFAR-100, and SVHN, with three orders of magnitude less overall compute. On CIFAR-10 we achieve a mean test error of 1.46%, which is a slight improvement upon the current state-of-the-art. The code for PBA is open source and is available at https://github.com/arcelien/pba.