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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper, we propose ByzSecAgg, an efficient secure aggregation schemefor federated learning that is protected against Byzantine attacks and privacyleakages. Processing individual updates to manage adversarial behavior, whilepreserving privacy of data against colluding nodes, requires some sort ofsecure secret sharing. However, the communication load for secret sharing oflong vectors of updates can be very high. ByzSecAgg solves this problem bypartitioning local updates into smaller sub-vectors and sharing them using rampsecret sharing. However, this sharing method does not admit bi-linearcomputations, such as pairwise distance calculations, needed byoutlier-detection algorithms. To overcome this issue, each user runs anotherround of ramp sharing, with different embedding of data in the sharingpolynomial. This technique, motivated by ideas from coded computing, enablessecure computation of pairwise distance. In addition, to maintain the integrityand privacy of the local update, ByzSecAgg also uses a vector commitmentmethod, in which the commitment size remains constant (i.e. does not increasewith the length of the local update), while simultaneously allowingverification of the secret sharing process. In terms of communication loads,ByzSecAgg significantly outperforms the state-of-the-art scheme, known as BREA.

ByzSecAgg: A Byzantine-Resistant Secure Aggregation Scheme for Federated Learning Based on Coded Computing and Vector Commitment