FedGrad: Optimisation in Decentralised Machine Learning

TL;DR

This paper introduces FedGrad, an adaptive optimization method for federated learning, demonstrating improved performance through experiments in a decentralized setting.

Contribution

The paper proposes a novel adaptive federated optimization algorithm called FedGrad and explores additional ideas to enhance federated learning performance.

Findings

FedGrad improves convergence speed in federated learning.

Experimental results show enhanced model accuracy.

The method outperforms existing federated optimization techniques.

Abstract

Federated Learning is a machine learning paradigm where we aim to train machine learning models in a distributed fashion. Many clients/edge devices collaborate with each other to train a single model on the central. Clients do not share their own datasets with each other, decoupling computation and data on the same device. In this paper, we propose yet another adaptive federated optimization method and some other ideas in the field of federated learning. We also perform experiments using these methods and showcase the improvement in the overall performance of federated learning.