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

Data heterogeneity, characterized by disparities in local data distributionacross clients, poses a significant challenge in federated learning.Substantial efforts have been devoted to addressing the heterogeneity in locallabel distribution. As minority classes suffer from worse accuracy due tooverfitting on local imbalanced data, prior methods often incorporateclass-balanced learning techniques during local training. Despite the improvedmean accuracy across all classes, we observe that empty classes-referring tocategories absent from a client's data distribution-are still not wellrecognized. This paper introduces FedED, a novel approach in heterogeneousfederated learning that integrates both empty-class distillation and logitsuppression simultaneously. Specifically, empty-class distillation leveragesknowledge distillation during local training on each client to retain essentialinformation related to empty classes from the global model. Moreover, logitsuppression directly penalizes network logits for non-label classes,effectively addressing misclassifications in minority classes that may bebiased toward majority classes. Extensive experiments validate the efficacy ofFedED, surpassing previous state-of-the-art methods across diverse datasetswith varying degrees of label distribution shift.

Not all Minorities are Equal: Empty-Class-Aware Distillation for Heterogeneous Federated Learning