Please use this identifier to cite or link to this item:
Title: Decentralized federated learning
Authors: Hitesh, Agarwal
Keywords: Engineering::Computer science and engineering
Issue Date: 2022
Publisher: Nanyang Technological University
Source: Hitesh, A. (2022). Decentralized federated learning. Final Year Project (FYP), Nanyang Technological University, Singapore.
Project: SCSE21-0200
Abstract: Conventional implementations of federated learning require a centralized entity to conduct and coordinate the training with a star communication architecture. However, this technique is prone to a single point of failure, e.g., when the central node is malicious. In this study, we explore decentralized federated learning frameworks where clients communicate with each other following a peer-to-peer mechanism rather than server-client. We study how communication topology and model partitioning affects the throughput and convergence metrics in decentralized federated learning. To make our study as practically applicable as possible, we include network link latencies in our performance metrics for a fair evaluation. Through our study, we conclude that the ring communication mechanism has the highest throughput with the best convergence performance metrics. In big networks, ring is almost 8 times as fast as centralized communications.
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
1.27 MBAdobe PDFView/Open

Page view(s)

Updated on May 16, 2022


Updated on May 16, 2022

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.