Header

UZH-Logo

Maintenance Infos

Fedstellar: A Platform for Training Models in a Privacy-preserving and Decentralized Fashion


Martínez Beltrán, Enrique Tomás; Sánchez Sánchez, Pedro Miguel; López Bernal, Sergio; Bovet, Gérôme; Gil Pérez, Manuel; Martínez Pérez, Gregorio; Celdrán, Alberto Huertas (2023). Fedstellar: A Platform for Training Models in a Privacy-preserving and Decentralized Fashion. In: Thirty-Second International Joint Conference on Artificial Intelligence {IJCAI-23}, Macau, SAR China, 19 August 2023 - 25 August 2023. IJCAI, 7154-7157.

Abstract

This paper presents Fedstellar, a platform for training decentralized Federated Learning (FL) models in heterogeneous topologies in terms of the number of federation participants and their connections. Fedstellar allows users to build custom topologies, enabling them to control the aggregation of model parameters in a decentralized manner. The platform offers a Web application for creating, managing, and connecting nodes to ensure data privacy and provides tools to measure, monitor, and analyze the performance of the nodes. The paper describes the functionalities of Fedstellar and its potential applications. To demonstrate the applicability of the platform, different use cases are presented in which decentralized, semi-decentralized, and centralized architectures are compared in terms of model performance, convergence time, and network overhead when collaboratively classifying hand-written digits using the MNIST dataset.

Abstract

This paper presents Fedstellar, a platform for training decentralized Federated Learning (FL) models in heterogeneous topologies in terms of the number of federation participants and their connections. Fedstellar allows users to build custom topologies, enabling them to control the aggregation of model parameters in a decentralized manner. The platform offers a Web application for creating, managing, and connecting nodes to ensure data privacy and provides tools to measure, monitor, and analyze the performance of the nodes. The paper describes the functionalities of Fedstellar and its potential applications. To demonstrate the applicability of the platform, different use cases are presented in which decentralized, semi-decentralized, and centralized architectures are compared in terms of model performance, convergence time, and network overhead when collaboratively classifying hand-written digits using the MNIST dataset.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Downloads

16 downloads since deposited on 12 Feb 2024
16 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Conference or Workshop Item (Paper), not_refereed, original work
Communities & Collections:03 Faculty of Economics > Department of Informatics
Dewey Decimal Classification:000 Computer science, knowledge & systems
Scopus Subject Areas:Physical Sciences > Artificial Intelligence
Scope:Discipline-based scholarship (basic research)
Language:English
Event End Date:25 August 2023
Deposited On:12 Feb 2024 13:50
Last Modified:14 Feb 2024 07:52
Publisher:IJCAI
Series Name:Proceedings of the International Joint Conference on Artificial Intelligence
ISBN:978-1-956792-03-4
OA Status:Hybrid
Publisher DOI:https://doi.org/10.24963/ijcai.2023/838
  • Content: Published Version
  • Language: English