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

Decentralized federated learning (DFL) is a variant of federated learning,where edge nodes only communicate with their one-hop neighbors to learn theoptimal model. However, as information exchange is restricted in a range ofone-hop in DFL, inefficient information exchange leads to more communicationrounds to reach the targeted training loss. This greatly reduces thecommunication efficiency. In this paper, we propose a new non-uniformquantization of model parameters to improve DFL convergence. Specifically, weapply the Lloyd-Max algorithm to DFL (LM-DFL) first to minimize thequantization distortion by adjusting the quantization levels adaptively.Convergence guarantee of LM-DFL is established without convex loss assumption.Based on LM-DFL, we then propose a new doubly-adaptive DFL, which jointlyconsiders the ascending number of quantization levels to reduce the amount ofcommunicated information in the training and adapts the quantization levels fornon-uniform gradient distributions. Experiment results based on MNIST andCIFAR-10 datasets illustrate the superiority of LM-DFL with the optimalquantized distortion and show that doubly-adaptive DFL can greatly improvecommunication efficiency.

Communication-Efficient Design for Quantized Decentralized Federated Learning