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

Deep neural network architectures are considered to be robust to randomperturbations. Nevertheless, it was shown that they could be severelyvulnerable to slight but carefully crafted perturbations of the input, termedas adversarial samples. In recent years, numerous studies have been conductedin this new area called "Adversarial Machine Learning" to devise newadversarial attacks and to defend against these attacks with more robust DNNarchitectures. However, almost all the research work so far has beenconcentrated on utilising model loss function to craft adversarial examples orcreate robust models. This study explores the usage of quantified epistemicuncertainty obtained from Monte-Carlo Dropout Sampling for adversarial attackpurposes by which we perturb the input to the areas where the model has notseen before. We proposed new attack ideas based on the epistemic uncertainty ofthe model. Our results show that our proposed hybrid attack approach increasesthe attack success rates from 82.59% to 85.40%, 82.86% to 89.92% and 88.06% to90.03% on MNIST Digit, MNIST Fashion and CIFAR-10 datasets, respectively.

Exploiting epistemic uncertainty of the deep learning models to generate adversarial samples