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

In black-box adversarial attacks, adversaries query the deep neural network(DNN), use the output to reconstruct gradients, and then optimize theadversarial inputs iteratively. In this paper, we study the method of addingwhite noise to the DNN output to mitigate such attacks, with a unique focus onthe trade-off analysis of noise level and query cost. The attacker's querycount (QC) is derived mathematically as a function of noise standard deviation.With this result, the defender can conveniently find the noise level needed tomitigate attacks for the desired security level specified by QC and limited DNNperformance loss. Our analysis shows that the added noise is drasticallymagnified by the small variation of DNN outputs, which makes the reconstructedgradient have an extremely low signal-to-noise ratio (SNR). Adding slight whitenoise with a standard deviation less than 0.01 is enough to increase QC by manyorders of magnitude without introducing any noticeable classification accuracyreduction. Our experiments demonstrate that this method can effectivelymitigate both soft-label and hard-label black-box attacks under realistic QCconstraints. We also show that this method outperforms many other defensemethods and is robust to the attacker's countermeasures.

Mitigating Black-Box Adversarial Attacks via Output Noise Perturbation