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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Recently, Convolutional Neural Networks (CNNs) demonstrate a considerablevulnerability to adversarial attacks, which can be easily misled by adversarialperturbations. With more aggressive methods proposed, adversarial attacks canbe also applied to the physical world, causing practical issues to various CNNpowered applications. To secure CNNs, adversarial attack detection isconsidered as the most critical approach. However, most existing works focus onsuperficial patterns and merely search a particular method to differentiate theadversarial inputs and natural inputs, ignoring the analysis of CNN innervulnerability. Therefore, they can only target to specific physical adversarialattacks, lacking expected versatility to different attacks. To address thisissue, we propose DoPa -- a comprehensive CNN detection methodology for variousphysical adversarial attacks. By interpreting the CNN's vulnerability, we findthat non-semantic adversarial perturbations can activate CNN with significantlyabnormal activations and even overwhelm other semantic input patterns'activations. Therefore, we add a self-verification stage to analyze thesemantics of distinguished activation patterns, which improves the CNNrecognition process. We apply such a detection methodology into both image andaudio CNN recognition scenarios. Experiments show that DoPa can achieve anaverage rate of 90% success for image attack detection and 92% success foraudio attack detection.  Announcement:[The original DoPa draft on arXiv was modified and submitted toa conference already, while this short abstract was submitted only for apresentation at the KDD 2019 AIoT Workshop.]

DoPa: A Comprehensive CNN Detection Methodology against Physical Adversarial Attacks