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

Backdoor attacks (BAs) are an emerging form of adversarial attack typicallyagainst deep neural network image classifiers. The attacker aims to have theclassifier learn to classify to a target class when test images from one ormore source classes contain a backdoor pattern, while maintaining high accuracyon all clean test images. Reverse-Engineering-based Defenses (REDs) against BAsdo not require access to the training set but only to an independent cleandataset. Unfortunately, most existing REDs rely on an unrealistic assumptionthat all classes except the target class are source classes of the attack. REDsthat do not rely on this assumption often require a large set of clean imagesand heavy computation. In this paper, we propose a Lagrangian-based RED (L-RED)that does not require knowledge of the number of source classes (or whether anattack is present). Our defense requires very few clean images to effectivelydetect BAs and is computationally efficient. Notably, we detect 56 out of 60BAs using only two clean images per class in our experiments on CIFAR-10.

L-RED: Efficient Post-Training Detection of Imperceptible Backdoor Attacks without Access to the Training Set