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

Most of the adversarial attack methods suffer from large perceptualdistortions such as visible artifacts, when the attack strength is relativelyhigh. These perceptual distortions contain a certain portion which contributesless to the attack success rate. This portion of distortions, which is inducedby unnecessary modifications and lack of proper perceptual distortionconstraint, is the target of the proposed framework. In this paper, we proposea perceptual distortion reduction framework to tackle this problem from twoperspectives. Firstly, we propose a perceptual distortion constraint and add itinto the objective function to jointly optimize the perceptual distortions andattack success rate. Secondly, we propose an adaptive penalty factor $\lambda$to balance the discrepancies between different samples. Since SGD andMomentum-SGD cannot optimize our complex non-convex problem, we exploit Adam inoptimization. Extensive experiments have verified the superiority of ourproposed framework.

A Perceptual Distortion Reduction Framework: Towards Generating Adversarial Examples with High Perceptual Quality and Attack Success Rate