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 consider a cross-silo federated learning (FL) setting where a machinelearning model with a fully connected first layer is trained between differentclients and a central server using FedAvg, and where the aggregation step canbe performed with secure aggregation (SA). We present SRATTA an attack relyingonly on aggregated models which, under realistic assumptions, (i) recovers datasamples from the different clients, and (ii) groups data samples coming fromthe same client together. While sample recovery has already been explored in anFL setting, the ability to group samples per client, despite the use of SA, isnovel. This poses a significant unforeseen security threat to FL andeffectively breaks SA. We show that SRATTA is both theoretically grounded andcan be used in practice on realistic models and datasets. We also proposecounter-measures, and claim that clients should play an active role toguarantee their privacy during training.

SRATTA : Sample Re-ATTribution Attack of Secure Aggregation in Federated Learning