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 the training of a model leveragingdecentralized data in client sites while preserving privacy by not collectingdata. However, one of the significant challenges of FL is limited computationand low communication bandwidth in resource limited edge client nodes. Toaddress this, several solutions have been proposed in recent times includingtransmitting sparse models and learning dynamic masks iteratively, amongothers. However, many of these methods rely on transmitting the model weightsthroughout the entire training process as they are based on ad-hoc or randompruning criteria. In this work, we propose Salient Grads, which simplifies theprocess of sparse training by choosing a data aware subnetwork before training,based on the model-parameter's saliency scores, which is calculated from thelocal client data. Moreover only highly sparse gradients are transmittedbetween the server and client models during the training process unlike mostmethods that rely on sharing the entire dense model in each round. We alsodemonstrate the efficacy of our method in a real world federated learningapplication and report improvement in wall-clock communication time.

SalientGrads: Sparse Models for Communication Efficient and Data Aware Distributed Federated Training