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

Heterogeneity of data distributed across clients limits the performance ofglobal models trained through federated learning, especially in the settingswith highly imbalanced class distributions of local datasets. In recent years,personalized federated learning (pFL) has emerged as a potential solution tothe challenges presented by heterogeneous data. However, existing pFL methodstypically enhance performance of local models at the expense of the globalmodel's accuracy. We propose FedHKD (Federated Hyper-Knowledge Distillation), anovel FL algorithm in which clients rely on knowledge distillation (KD) totrain local models. In particular, each client extracts and sends to the serverthe means of local data representations and the corresponding soft predictions-- information that we refer to as ``hyper-knowledge". The server aggregatesthis information and broadcasts it to the clients in support of local training.Notably, unlike other KD-based pFL methods, FedHKD does not rely on a publicdataset nor it deploys a generative model at the server. We analyze convergenceof FedHKD and conduct extensive experiments on visual datasets in a variety ofscenarios, demonstrating that FedHKD provides significant improvement in bothpersonalized as well as global model performance compared to state-of-the-artFL methods designed for heterogeneous data settings.

The Best of Both Worlds: Accurate Global and Personalized Models through Federated Learning with Data-Free Hyper-Knowledge Distillation