DELIGHT
IFIP
IEEE

DELIGHT

ADVANCING FEDERATED LEARNING WHILE REDUCING THE CARBON FOOTPRINT

News


20/09/2022

The project start officially on 28 April 2023.


20/11/2022

Kick-off meeting at the University of Avignon on 6-7 January 2023:


10/04/2023:

Extension of the project by six months due to delays in recruiting PhD students. The project is scheduled to end on 28/04/2027


01/04/2024

May 06 2024, Tutorial presented by Georges Da Costa on techniques used to measure energy consumption given for Phd students involved in the project.


24/07/2024

Best Poster Award ICDCS24: Optimal Variance-Reduced Client Sampling for Multiple Models Federated Learning Haoran Zhang, Zekai Li, Zejun Gong, Marie Siew, Carlee Joe-Wong, Rachid El-Azouzi


10/09/2024

Physical meeting at IRIT, University of Toulouse on 6 and 7 November 2024


10/10/2024

First tools developed by the project to measure the energy consumption of federated learning (FL) experiments based on Flower on the Grid'5000 (G5K) testbed.


05/05/2025

Physical meeting at CNRS/LAAS, on 15 and 16 July 2025


25/06/2025

FedEator, the energy measurement framework for FL experiments, was officially introduced (1st time) at COMPAS 2025, Bordeaux


08/07/2025

FedEator was presented at the IRIT PhD Days 2025, Toulouse