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

Scaling up the convolutional neural network (CNN) size (e.g., width, depth,etc.) is known to effectively improve model accuracy. However, the large modelsize impedes training on resource-constrained edge devices. For instance,federated learning (FL) may place undue burden on the compute capability ofedge nodes, even though there is a strong practical need for FL due to itsprivacy and confidentiality properties. To address the resource-constrainedreality of edge devices, we reformulate FL as a group knowledge transfertraining algorithm, called FedGKT. FedGKT designs a variant of the alternatingminimization approach to train small CNNs on edge nodes and periodicallytransfer their knowledge by knowledge distillation to a large server-side CNN.FedGKT consolidates several advantages into a single framework: reduced demandfor edge computation, lower communication bandwidth for large CNNs, andasynchronous training, all while maintaining model accuracy comparable toFedAvg. We train CNNs designed based on ResNet-56 and ResNet-110 using threedistinct datasets (CIFAR-10, CIFAR-100, and CINIC-10) and their non-I.I.D.variants. Our results show that FedGKT can obtain comparable or even slightlyhigher accuracy than FedAvg. More importantly, FedGKT makes edge trainingaffordable. Compared to the edge training using FedAvg, FedGKT demands 9 to 17times less computational power (FLOPs) on edge devices and requires 54 to 105times fewer parameters in the edge CNN. Our source code is released at FedML(https://fedml.ai).

Group Knowledge Transfer: Federated Learning of Large CNNs at the Edge