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

While machine learning (ML) models are being increasingly trusted to makedecisions in different and varying areas, the safety of systems using suchmodels has become an increasing concern. In particular, ML models are oftentrained on data from potentially untrustworthy sources, providing adversarieswith the opportunity to manipulate them by inserting carefully crafted samplesinto the training set. Recent work has shown that this type of attack, called apoisoning attack, allows adversaries to insert backdoors or trojans into themodel, enabling malicious behavior with simple external backdoor triggers atinference time and only a blackbox perspective of the model itself. Detectingthis type of attack is challenging because the unexpected behavior occurs onlywhen a backdoor trigger, which is known only to the adversary, is present.Model users, either direct users of training data or users of pre-trained modelfrom a catalog, may not guarantee the safe operation of their ML-based system.In this paper, we propose a novel approach to backdoor detection and removalfor neural networks. Through extensive experimental results, we demonstrate itseffectiveness for neural networks classifying text and images. To the best ofour knowledge, this is the first methodology capable of detecting poisonousdata crafted to insert backdoors and repairing the model that does not requirea verified and trusted dataset.

Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering