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

Deep neural networks, like many other machine learning models, have recentlybeen shown to lack robustness against adversarially crafted inputs. Theseinputs are derived from regular inputs by minor yet carefully selectedperturbations that deceive machine learning models into desiredmisclassifications. Existing work in this emerging field was largely specificto the domain of image classification, since the high-entropy of images can beconveniently manipulated without changing the images' overall visualappearance. Yet, it remains unclear how such attacks translate to moresecurity-sensitive applications such as malware detection - which may posesignificant challenges in sample generation and arguably grave consequences forfailure.  In this paper, we show how to construct highly-effective adversarial samplecrafting attacks for neural networks used as malware classifiers. Theapplication domain of malware classification introduces additional constraintsin the adversarial sample crafting problem when compared to the computer visiondomain: (i) continuous, differentiable input domains are replaced by discrete,often binary inputs; and (ii) the loose condition of leaving visual appearanceunchanged is replaced by requiring equivalent functional behavior. Wedemonstrate the feasibility of these attacks on many different instances ofmalware classifiers that we trained using the DREBIN Android malware data set.We furthermore evaluate to which extent potential defensive mechanisms againstadversarial crafting can be leveraged to the setting of malware classification.While feature reduction did not prove to have a positive impact, distillationand re-training on adversarially crafted samples show promising results.

Adversarial Perturbations Against Deep Neural Networks for Malware Classification