Accelerated Training via Device Similarity in Federated Learning

Wang, Yuanli; Wolfrath, Joel; Sreekumar, Nikhil; Kumar, Dharmender; Chandra, Abhishek

doi:10.1145/3434770.3459734

Cited by 13 publications

(2 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The focus of this work is on bandwidth optimization rather than the speed of accuracy convergence. Although not directly connected with the scope of our investigation, but still very relevant to the work of Wang et al [26]. In their research, they propose to group the devices based on their data similarity, followed by selecting the devices with the best performance capacity.…”

Section: Related Workmentioning

confidence: 97%

Impact of Network Topology on the Convergence of Decentralized Federated Learning Systems

Kavalionak

Carlini

Dazzi

et al. 2021

2021 IEEE Symposium on Computers and Communications (ISCC)

View full text Add to dashboard Cite

Federated learning is a popular framework that enables harvesting edge resources' computational power to train a machine learning model distributively. However, it is not always feasible or profitable to have a centralized server that controls and synchronizes the training process. In this paper, we consider the problem of training a machine learning model over a network of nodes in a fully decentralized fashion. In particular, we look for empirical evidence on how sensitive is the training process for various network characteristics and communication parameters. We present the outcome of several simulations conducted with different network topologies, datasets, and machine learning models.

show abstract