Enabling Data Diversity: Efficient Automatic Augmentation via Regularized Adversarial Training

03/30/2021
by   Yunhe Gao, et al.
16

Data augmentation has proved extremely useful by increasing training data variance to alleviate overfitting and improve deep neural networks' generalization performance. In medical image analysis, a well-designed augmentation policy usually requires much expert knowledge and is difficult to generalize to multiple tasks due to the vast discrepancies among pixel intensities, image appearances, and object shapes in different medical tasks. To automate medical data augmentation, we propose a regularized adversarial training framework via two min-max objectives and three differentiable augmentation models covering affine transformation, deformation, and appearance changes. Our method is more automatic and efficient than previous automatic augmentation methods, which still rely on pre-defined operations with human-specified ranges and costly bi-level optimization. Extensive experiments demonstrated that our approach, with less training overhead, achieves superior performance over state-of-the-art auto-augmentation methods on both tasks of 2D skin cancer classification and 3D organs-at-risk segmentation.

READ FULL TEXT

page 9

page 11

research
01/24/2023

Data Augmentation Alone Can Improve Adversarial Training

Adversarial training suffers from the issue of robust overfitting, which...
research
07/21/2022

Auto Machine Learning for Medical Image Analysis by Unifying the Search on Data Augmentation and Neural Architecture

Automated data augmentation, which aims at engineering augmentation poli...
research
02/06/2023

GAT: Guided Adversarial Training with Pareto-optimal Auxiliary Tasks

While leveraging additional training data is well established to improve...
research
05/16/2018

Adversarial Training for Patient-Independent Feature Learning with IVOCT Data for Plaque Classification

Deep learning methods have shown impressive results for a variety of med...
research
04/13/2023

Understanding Overfitting in Adversarial Training via Kernel Regression

Adversarial training and data augmentation with noise are widely adopted...
research
03/18/2021

TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation

Automatic augmentation methods have recently become a crucial pillar for...
research
09/02/2020

Evaluation of Deep Convolutional Generative Adversarial Networks for data augmentation of chest X-ray images

Medical image datasets are usually imbalanced, due to the high costs of ...

Please sign up or login with your details

Forgot password? Click here to reset