Fedstellar: A Platform for Training Models in a Privacy-Preserving and Decentralized Fashion
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.
Cite
Text
Beltrán et al. "Fedstellar: A Platform for Training Models in a Privacy-Preserving and Decentralized Fashion." International Joint Conference on Artificial Intelligence, 2023. doi:10.24963/IJCAI.2023/838Markdown
[Beltrán et al. "Fedstellar: A Platform for Training Models in a Privacy-Preserving and Decentralized Fashion." International Joint Conference on Artificial Intelligence, 2023.](https://mlanthology.org/ijcai/2023/beltran2023ijcai-fedstellar/) doi:10.24963/IJCAI.2023/838BibTeX
@inproceedings{beltran2023ijcai-fedstellar,
title = {{Fedstellar: A Platform for Training Models in a Privacy-Preserving and Decentralized Fashion}},
author = {Beltrán, Enrique Tomás Martínez and Sánchez, Pedro Miguel Sánchez and Bernal, Sergio López and Bovet, Gérôme and Pérez, Manuel Gil and Pérez, Gregorio Martínez and Celdrán, Alberto Huertas},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2023},
pages = {7154-7157},
doi = {10.24963/IJCAI.2023/838},
url = {https://mlanthology.org/ijcai/2023/beltran2023ijcai-fedstellar/}
}