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

Personalized federated learning is proposed to handle the data heterogeneityproblem amongst clients by learning dedicated tailored local models for eachuser. However, existing works are often built in a centralized way, leading tohigh communication pressure and high vulnerability when a failure or an attackon the central server occurs. In this work, we propose a novel personalizedfederated learning framework in a decentralized (peer-to-peer) communicationprotocol named Dis-PFL, which employs personalized sparse masks to customizesparse local models on the edge. To further save the communication andcomputation cost, we propose a decentralized sparse training technique, whichmeans that each local model in Dis-PFL only maintains a fixed number of activeparameters throughout the whole local training and peer-to-peer communicationprocess. Comprehensive experiments demonstrate that Dis-PFL significantly savesthe communication bottleneck for the busiest node among all clients and, at thesame time, achieves higher model accuracy with less computation cost andcommunication rounds. Furthermore, we demonstrate that our method can easilyadapt to heterogeneous local clients with varying computation complexities andachieves better personalized performances.

DisPFL: Towards Communication-Efficient Personalized Federated Learning via Decentralized Sparse Training