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 algorithms have demonstrated a strong ability indetecting Android malware, they can be evaded by sparse evasion attacks craftedby injecting a small set of fake components, e.g., permissions and systemcalls, without compromising intrusive functionality. Previous work has shownthat, to improve robustness against such attacks, learning algorithms shouldavoid overemphasizing few discriminant features, providing instead decisionsthat rely upon a large subset of components. In this work, we investigatewhether gradient-based attribution methods, used to explain classifiers'decisions by identifying the most relevant features, can be used to helpidentify and select more robust algorithms. To this end, we propose to exploittwo different metrics that represent the evenness of explanations, and a newcompact security measure called Adversarial Robustness Metric. Our experimentsconducted on two different datasets and five classification algorithms forAndroid malware detection show that a strong connection exists between theuniformity of explanations and adversarial robustness. In particular, we foundthat popular techniques like Gradient*Input and Integrated Gradients arestrongly correlated to security when applied to both linear and nonlineardetectors, while more elementary explanation techniques like the simpleGradient do not provide reliable information about the robustness of suchclassifiers.

Do Gradient-based Explanations Tell Anything About Adversarial Robustness to Android Malware?