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 is a distributed collaborative machine learning paradigmthat has gained strong momentum in recent years. In federated learning, acentral server periodically coordinates models with clients and aggregates themodels trained locally by clients without necessitating access to local data.Despite its potential, the implementation of federated learning continues toencounter several challenges, predominantly the slow convergence that islargely due to data heterogeneity. The slow convergence becomes particularlyproblematic in cross-device federated learning scenarios where clients may bestrongly limited by computing power and storage space, and hence counteractingmethods that induce additional computation or memory cost on the client sidesuch as auxiliary objective terms and larger training iterations can beimpractical. In this paper, we propose a novel federated aggregation strategy,TurboSVM-FL, that poses no additional computation burden on the client side andcan significantly accelerate convergence for federated classification task,especially when clients are "lazy" and train their models solely for few epochsfor next global aggregation. TurboSVM-FL extensively utilizes support vectormachine to conduct selective aggregation and max-margin spread-outregularization on class embeddings. We evaluate TurboSVM-FL on multipledatasets including FEMNIST, CelebA, and Shakespeare using user-independentvalidation with non-iid data distribution. Our results show that TurboSVM-FLcan significantly outperform existing popular algorithms on convergence rateand reduce communication rounds while delivering better test metrics includingaccuracy, F1 score, and MCC.

TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients