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

Federated Learning (FL) enables distributed clients to collaboratively trainmodels without exposing their private data. However, it is difficult toimplement efficient FL due to limited resources. Most existing works compressthe transmitted gradients or prune the global model to reduce the resourcecost, but leave the compressed or pruned parameters under-optimized, whichdegrades the training performance. To address this issue, the neuralcomposition technique constructs size-adjustable models by composing low-ranktensors, allowing every parameter in the global model to learn the knowledgefrom all clients. Nevertheless, some tensors can only be optimized by a smallfraction of clients, thus the global model may get insufficient training,leading to a long completion time, especially in heterogeneous edge scenarios.To this end, we enhance the neural composition technique, enabling allparameters to be fully trained. Further, we propose a lightweight FL framework,called Heroes, with enhanced neural composition and adaptive local update. Agreedy-based algorithm is designed to adaptively assign the proper tensors andlocal update frequencies for participating clients according to theirheterogeneous capabilities and resource budgets. Extensive experimentsdemonstrate that Heroes can reduce traffic consumption by about 72.05\% andprovide up to 2.97$\times$ speedup compared to the baselines.

Heroes: Lightweight Federated Learning with Neural Composition and Adaptive Local Update in Heterogeneous Edge Networks