Abstract:Data augmentation (DA) has been widely used to improve the generalization of deep neural networks. While existing DA methods have proven effective, they often rely on augmentation operations with random magnitudes to each sample. However, this approach can inadvertently introduce noise, induce distribution shifts, and increase the risk of overfitting. In this paper, we propose EntAugment, a tuning-free and adaptive DA framework. Unlike previous work, EntAugment dynamically assesses and adjusts the augmentation magnitudes for each sample during training, leveraging insights into both the inherent complexities of training samples and the evolving status of deep models. Specifically, in EntAugment, the magnitudes are determined by the information entropy derived from the probability distribution obtained by applying the softmax function to the model's output. In addition, to further enhance the efficacy of EntAugment, we introduce a novel entropy regularization term, EntLoss, which complements the EntAugment approach. Theoretical analysis further demonstrates that EntLoss, compared to traditional cross-entropy loss, achieves closer alignment between the model distributions and underlying dataset distributions. Moreover, EntAugment and EntLoss can be utilized separately or jointly. We conduct extensive experiments across multiple image classification tasks and network architectures with thorough comparisons of existing DA methods. Importantly, the proposed methods outperform others without introducing any auxiliary models or noticeable extra computational costs, highlighting both effectiveness and efficiency. Code is available at <a class="link-external link-https" href="https://github.com/Jackbrocp/EntAugment" rel="external noopener nofollow">this https URL</a>.

Optimizing distortion magnitude for data augmentation in few-shot remote sensing scene classification

Optimizing Few-Shot Remote Sensing Scene Classification Based on an Improved Data Augmentation Approach

Boosting Unsupervised Contrastive Learning Using Diffusion-Based Data Augmentation from Scratch

DualAug: Exploiting Additional Heavy Augmentation with OOD Data Rejection

Data augmentation and few-shot change detection in forest remote sensing

The Performance Research of the Data Augmentation Method for Image Classification

See Better Before Looking Closer: Weakly Supervised Data Augmentation Network for Fine-Grained Visual Classification

A Novel Multi-Sample Data Augmentation Method for Oriented Object Detection in Remote Sensing Images

Data Augmentation Revisited: Rethinking the Distribution Gap between Clean and Augmented Data

A Preliminary Study on Data Augmentation of Deep Learning for Image Classification

A Review of Data Augmentation Methods of Remote Sensing Image Target Recognition

A Closer Look at Data Augmentation Strategies for Finetuning-Based Low/Few-Shot Object Detection

EntAugment: Entropy-Driven Adaptive Data Augmentation Framework for Image Classification

Data Augmentation for Low-Level Vision: CutBlur and Mixture-of-Augmentation

Leveraging Smooth Deformation Augmentation for LiDAR Point Cloud Semantic Segmentation

A class distribution learning method for few-shot remote sensing scene classification

Tied-Augment: Controlling Representation Similarity Improves Data Augmentation

Learning Optimal Data Augmentation Policies via Bayesian Optimization for Image Classification Tasks

A Deep Ordinal Distortion Estimation Approach for Distortion Rectification

Effective Data Augmentation With Diffusion Models