Please use this identifier to cite or link to this item:
Title: Incentivizing Participation to Distributed Neural Network Training
Authors: Nikolaidis, Spyridon
Refanidis, Ioannis
Editors: Iliadis, Lazaros
Macintyre, John
Jayne, Chrisina
Pimenidis, Elias
Type: Conference Paper
Subjects: FRASCATI::Natural sciences::Computer and information sciences
Keywords: Decentralized neural network training
Distributed Ledger Technology
Smart Contracts
Issue Date: 1-Jul-2021
Publisher: Springer
First Page: 364
Last Page: 374
Volume Title: Proceedings of the 22nd Engineering Applications of Neural Networks Conference
Part of Series: Proceedings of the International Neural Networks Society
Part of Series: Proceedings of the International Neural Networks Society
Abstract: During the last years a vast number of online sensors continuously generate data that can be utilized to create novel deep learning applications. Training very large models requires enormous processing power; thus, the evident way to follow is to lease the power of a corporate data center. But the diffusion of Artificial Intelligence to an always increasing number of human activities, constantly attracts new researchers who wish to train and test their models. Our work on LEARNAE is a proposal for a purely distributed neural network training, based on a peer-to-peer and permissionless architecture. LEARNAE allows individual researchers to join forces, in order to collaboratively train a model. The process utilizes modern Distributed Ledger Technology and it is fully democratized, prioritizing decentralization, fault tolerance and privacy. In this paper we add another piece to the puzzle: A method for incentivizing peers to participate to the training swarm, even if they don’t have any interest in the produced neural network. This is achieved by embedding a reward subsystem to LEARNAE; thus, peers who contribute to teamwork can receive a proportional digital payment.
ISBN: 978-3-030-80567-8
Electronic ISBN: 978-3-030-80568-5
Other Identifiers: 10.1007/978-3-030-80568-5_30
Appears in Collections:Department of Applied Informatics

Files in This Item:
File Description SizeFormat 
eann2021_attachment_15.pdfpostprint579,32 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.