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

We propose a numerical accountant for evaluating the tight$(\varepsilon,\delta)$-privacy loss for algorithms with discrete onedimensional output. The method is based on the privacy loss distributionformalism and it uses the recently introduced fast Fourier transform basedaccounting technique. We carry out an error analysis of the method in terms ofmoment bounds of the privacy loss distribution which leads to rigorous lowerand upper bounds for the true $(\varepsilon,\delta)$-values. As an application,we present a novel approach to accurate privacy accounting of the subsampledGaussian mechanism. This completes the previously proposed analysis by givingstrict lower and upper bounds for the privacy parameters. We demonstrate theperformance of the accountant on the binomial mechanism and show that ourapproach allows decreasing noise variance up to 75 percent at equal privacycompared to existing bounds in the literature. We also illustrate how tocompute tight bounds for the exponential mechanism applied to counting queries.

Tight Differential Privacy for Discrete-Valued Mechanisms and for the Subsampled Gaussian Mechanism Using FFT