ReSmooth: Detecting and Utilizing OOD Samples when Training with Data Augmentation

TL;DR

ReSmooth is a framework that detects out-of-distribution augmented samples during training and leverages them with different labels, improving the effectiveness of data augmentation techniques in deep neural network training.

Contribution

It introduces a method to identify and utilize OOD samples in augmented data, enhancing data augmentation strategies for better model performance.

Findings

ReSmooth improves classification accuracy across multiple benchmarks.

The framework effectively detects OOD samples using Gaussian mixture models.

ReSmooth enhances existing augmentation methods like RandAugment, rotate, and jigsaw.

Abstract

Data augmentation (DA) is a widely used technique for enhancing the training of deep neural networks. Recent DA techniques which achieve state-of-the-art performance always meet the need for diversity in augmented training samples. However, an augmentation strategy that has a high diversity usually introduces out-of-distribution (OOD) augmented samples and these samples consequently impair the performance. To alleviate this issue, we propose ReSmooth, a framework that firstly detects OOD samples in augmented samples and then leverages them. To be specific, we first use a Gaussian mixture model to fit the loss distribution of both the original and augmented samples and accordingly split these samples into in-distribution (ID) samples and OOD samples. Then we start a new training where ID and OOD samples are incorporated with different smooth labels. By treating ID samples and OOD samples unequally, we can make better use of the diverse augmented data. Further, we incorporate our ReSmooth framework with negative data augmentation strategies. By properly handling their intentionally created OOD samples, the classification performance of negative data augmentations is largely ameliorated. Experiments on several classification benchmarks show that ReSmooth can be easily extended to existing augmentation strategies (such as RandAugment, rotate, and jigsaw) and improve on them. Our code is available at https://github.com/Chenyang4/ReSmooth.