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

Over the past decade, side-channels have proven to be significant andpractical threats to modern computing systems. Recent attacks have allexploited the underlying shared hardware. While practical, mounting such acomplicated attack is still akin to listening on a private conversation in acrowded train station. The attacker has to either perform significant manuallabor or use AI systems to automate the process. The recent academic literaturepoints to the latter option. With the abundance of cheap computing power andthe improvements made in AI, it is quite advantageous to automate such tasks.By using AI systems however, malicious parties also inherit their weaknesses.One such weakness is undoubtedly the vulnerability to adversarial samples.  In contrast to the previous literature, for the first time, we propose theuse of adversarial learning as a defensive tool to obfuscate and mask privateinformation. We demonstrate the viability of this approach by first trainingCNNs and other machine learning classifiers on leakage trace of differentprocesses. After training highly accurate models (99+% accuracy), weinvestigate their resolve against adversarial learning methods. By applyingminimal perturbations to input traces, the adversarial traffic by the defendercan run as an attachment to the original process and cloak it against amalicious classifier.  Finally, we investigate whether an attacker can protect her classifier modelby employing adversarial defense methods, namely adversarial re-training anddefensive distillation. Our results show that even in the presence of anintelligent adversary that employs such techniques, all 10 of the testedadversarial learning methods still manage to successfully craft adversarialperturbations and the proposed cloaking methodology succeeds.

DeepCloak: Adversarial Crafting As a Defensive Measure to Cloak Processes