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 (Trojan) attack is a common threat to deep neural networks, wheresamples from one or more source classes embedded with a backdoor trigger willbe misclassified to adversarial target classes. Existing methods for detectingwhether a classifier is backdoor attacked are mostly designed for attacks witha single adversarial target (e.g., all-to-one attack). To the best of ourknowledge, without supervision, no existing methods can effectively address themore general X2X attack with an arbitrary number of source classes, each pairedwith an arbitrary target class. In this paper, we propose UMD, the firstUnsupervised Model Detection method that effectively detects X2X backdoorattacks via a joint inference of the adversarial (source, target) class pairs.In particular, we first define a novel transferability statistic to measure andselect a subset of putative backdoor class pairs based on a proposed clusteringapproach. Then, these selected class pairs are jointly assessed based on anaggregation of their reverse-engineered trigger size for detection inference,using a robust and unsupervised anomaly detector we proposed. We conductcomprehensive evaluations on CIFAR-10, GTSRB, and Imagenette dataset, and showthat our unsupervised UMD outperforms SOTA detectors (even with supervision) by17%, 4%, and 8%, respectively, in terms of the detection accuracy againstdiverse X2X attacks. We also show the strong detection performance of UMDagainst several strong adaptive attacks.

UMD: Unsupervised Model Detection for X2X Backdoor Attacks