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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In recent years, there has been growing concern over the vulnerability ofconvolutional neural networks (CNNs) to image perturbations. However, achievinggeneral robustness against different types of perturbations remainschallenging, in which enhancing robustness to some perturbations (e.g.,adversarial perturbations) may degrade others (e.g., common corruptions). Inthis paper, we demonstrate that adversarial training with an emphasis on phasecomponents significantly improves model performance on clean, adversarial, andcommon corruption accuracies. We propose a frequency-based data augmentationmethod, Adversarial Amplitude Swap, that swaps the amplitude spectrum betweenclean and adversarial images to generate two novel training images: adversarialamplitude and adversarial phase images. These images act as substitutes foradversarial images and can be implemented in various adversarial trainingsetups. Through extensive experiments, we demonstrate that our method enablesthe CNNs to gain general robustness against different types of perturbationsand results in a uniform performance against all types of common corruptions.

Exploiting Frequency Spectrum of Adversarial Images for General Robustness