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/

Zendy Plus

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

Zendy Tools

Zendy Open

Security-sensitive applications that rely on Deep Neural Networks (DNNs) arevulnerable to small perturbations that are crafted to generate AdversarialExamples(AEs). The AEs are imperceptible to humans and cause DNN to misclassifythem. Many defense and detection techniques have been proposed. Model'sconfidences and Dropout, as a popular way to estimate the model's uncertainty,have been used for AE detection but they showed limited success against black-and gray-box attacks. Moreover, the state-of-the-art detection techniques havebeen designed for specific attacks or broken by others, need knowledge aboutthe attacks, are not consistent, increase model parameters overhead, aretime-consuming, or have latency in inference time. To trade off these factors,we revisit the model's uncertainty and confidences and propose a novelunsupervised ensemble AE detection mechanism that 1) uses the uncertaintymethod called SelectiveNet, 2) processes model layers outputs, i.e.featuremaps, to generate new confidence probabilities. The detection method is calledSelective and Feature based Adversarial Detection (SFAD). Experimental resultsshow that the proposed approach achieves better performance against black- andgray-box attacks than the state-of-the-art methods and achieves comparableperformance against white-box attacks. Moreover, results show that SFAD isfully robust against High Confidence Attacks (HCAs) for MNIST and partiallyrobust for CIFAR10 datasets.

Revisiting Model's Uncertainty and Confidences for Adversarial Example Detection