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

Federated Learning, as a popular paradigm for collaborative training, isvulnerable against privacy attacks. Different privacy levels regarding users'attitudes need to be satisfied locally, while a strict privacy guarantee forthe global model is also required centrally. Personalized Local DifferentialPrivacy (PLDP) is suitable for preserving users' varying local privacy, yetonly provides a central privacy guarantee equivalent to the worst-case localprivacy level. Thus, achieving strong central privacy as well as personalizedlocal privacy with a utility-promising model is a challenging problem. In thiswork, a general framework (APES) is built up to strengthen model privacy underpersonalized local privacy by leveraging the privacy amplification effect ofthe shuffle model. To tighten the privacy bound, we quantify the heterogeneouscontributions to the central privacy user by user. The contributions arecharacterized by the ability of generating "echos" from the perturbation ofeach user, which is carefully measured by proposed methods Neighbor Divergenceand Clip-Laplace Mechanism. Furthermore, we propose a refined framework(S-APES) with the post-sparsification technique to reduce privacy loss inhigh-dimension scenarios. To the best of our knowledge, the impact of shufflingon personalized local privacy is considered for the first time. We provide astrong privacy amplification effect, and the bound is tighter than the baselineresult based on existing methods for uniform local privacy. Experimentsdemonstrate that our frameworks ensure comparable or higher accuracy for theglobal model.

Echo of Neighbors: Privacy Amplification for Personalized Private Federated Learning with Shuffle Model