Safe Augmentation: Learning Task-Specific Transformations from Data

Abstract

Data augmentation is widely used as a part of the training process applied to deep learning models, especially in the computer vision domain. Currently, common data augmentation techniques are designed manually. Therefore they require expert knowledge and time. Moreover, augmentations are dataset-specific, and the optimal augmentations set on a specific dataset has limited transferability to others. We present a simple and explainable method called $\text{Safe Augmentation}$ that can learn task-specific data augmentation techniques that do not change the data distribution and improve the generalization of the model. We propose to use safe augmentation in two ways: for model fine-tuning and along with other augmentation techniques. Our method is model-agnostic, easy to implement, and achieves better accuracy on CIFAR-10, CIFAR-100, SVHN, Tiny ImageNet, and Cityscapes datasets comparing to baseline augmentation techniques. The code is available at $\text{href}$.

References in corpus (16)

• Adam: A Method for Stochastic Optimization
• Improved Regularization of Convolutional Neural Networks with Cutout
• The Effectiveness of Data Augmentation in Image Classification using Deep Learning
• Densely Connected Convolutional Networks
• The Cityscapes Dataset for Semantic Urban Scene Understanding
• Data Augmentation Generative Adversarial Networks
• AutoAugment: Learning Augmentation Policies from Data
• Feature Pyramid Networks for Object Detection
• Smart Augmentation - Learning an Optimal Data Augmentation Strategy
• The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches
• Some Improvements on Deep Convolutional Neural Network Based Image Classification
• GAN Augmentation: Augmenting Training Data using Generative Adversarial Networks
• Identity Mappings in Deep Residual Networks
• A Bayesian Data Augmentation Approach for Learning Deep Models
• Data augmentation instead of explicit regularization
• Conditional Infilling GANs for Data Augmentation in Mammogram Classification