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

Trustworthy Federated Learning (TFL) typically leverages protectionmechanisms to guarantee privacy. However, protection mechanisms inevitablyintroduce utility loss or efficiency reduction while protecting data privacy.Therefore, protection mechanisms and their parameters should be carefullychosen to strike an optimal tradeoff between \textit{privacy leakage},\textit{utility loss}, and \textit{efficiency reduction}. To this end,federated learning practitioners need tools to measure the three factors andoptimize the tradeoff between them to choose the protection mechanism that ismost appropriate to the application at hand. Motivated by this requirement, wepropose a framework that (1) formulates TFL as a problem of finding aprotection mechanism to optimize the tradeoff between privacy leakage, utilityloss, and efficiency reduction and (2) formally defines bounded measurements ofthe three factors. We then propose a meta-learning algorithm to approximatethis optimization problem and find optimal protection parameters forrepresentative protection mechanisms, including Randomization, HomomorphicEncryption, Secret Sharing, and Compression. We further design estimationalgorithms to quantify these found optimal protection parameters in a practicalhorizontal federated learning setting and provide a theoretical analysis of theestimation error.

A Meta-learning Framework for Tuning Parameters of Protection Mechanisms in Trustworthy Federated Learning