English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Adversarial training is exploited to develop a robust Deep Neural Network(DNN) model against the malicious altered data. These attacks may havecatastrophic effects on DNN models but are indistinguishable for a human being.For example, an external attack can modify an image adding noises invisible fora human eye, but a DNN model misclassified the image. A key objective fordeveloping robust DNN models is to use a learning algorithm that is fast butcan also give model that is robust against different types of adversarialattacks. Especially for adversarial training, enormously long training timesare needed for obtaining high accuracy under many different types ofadversarial samples generated using different adversarial attack techniques.  This paper aims at accelerating the adversarial training to enable fastdevelopment of robust DNN models against adversarial attacks. The generalmethod for improving the training performance is the hyperparametersfine-tuning, where the learning rate is one of the most crucialhyperparameters. By modifying its shape (the value over time) and value duringthe training, we can obtain a model robust to adversarial attacks faster thanstandard training.  First, we conduct experiments on two different datasets (CIFAR10, CIFAR100),exploring various techniques. Then, this analysis is leveraged to develop anovel fast training methodology, AccelAT, which automatically adjusts thelearning rate for different epochs based on the accuracy gradient. Theexperiments show comparable results with the related works, and in severalexperiments, the adversarial training of DNNs using our AccelAT framework isconducted up to 2 times faster than the existing techniques. Thus, our findingsboost the speed of adversarial training in an era in which security andperformance are fundamental optimization objectives in DNN-based applications.

AccelAT: A Framework for Accelerating the Adversarial Training of Deep Neural Networks Through Accuracy Gradient