ai4-metadata.yml

metadata_version: 2.0.0
title: Federated learning server
summary: Federated learning server with Flower.
description: |-
  Federated Learning server deployed using [Flower](https://flower.ai/).
  
  Run your federated learning (FL) application from scratch with the server running on AI4EOSC.
  The FL clients can be deployed in external cloud machines, locally in their own workstations, or on the AI4EOSC platform (accessing to GPUs and other computing resources).
  With this tool you can train global machine/deep learning model on different clients in a distributed way following a federated learning architecture.
  
  Customize the number of rounds the FL training is performed, the aggregation strategy used, the minimum number of clients, and the error metric to be used among other settings.

  <img class='fit', src='https://www.researchgate.net/publication/362088869/figure/fig2/AS:1179390767575052@1658200079112/Scheme-of-a-federated-learning-approach.png'/>
  
dates:
  created: '2023-06-28'
  updated: '2024-08-12'
links:
  source_code: https://github.com/ai4os/ai4os-federated-server
  docker_image: ai4oshub/ai4os-federated-server
  ai4_template: ai4-template/1.9.9
tags:
  - federated learning
tasks:
  - Other
categories:
  - AI4 tools
libraries:
  - Other
data-type:
  - Other