BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Personalized Federated Learning
DTSTART:20240223T110000
DTEND:20240223T120000
DTSTAMP:20260429T152737Z
UID:e9204caf841110d21c6abfce0a84c00c69f398e8eae3ce1fcbc1fbc8
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Giovanni Neglia\,\n\nInria\, France.\n\n \nAbstract:\nFed
 erated Learning (FL) has emerged as a prominent methodology for collective
 ly training a shared machine learning model over the Internet while mainta
 ining data at client endpoints. However\, due to variations in data distri
 butions among clients\, the efficacy of the shared model may be limited fo
 r certain clients\, prompting the exploration of FL techniques aimed at tr
 aining personalized models.\nThis talk will provide a succinct overview of
  three recent FL approaches\, each grounded in distinct assumptions and ac
 companied by corresponding learning algorithms: (1) kNNPer\, which leverag
 es a shared representation\; (2) FedEM\, which assumes local client distri
 butions as mixtures of a common set of underlying distributions\; and (3) 
 ColME\, which operates under the assumption of shared distributions across
  clusters of clients.\n\nReferences:\n- Marfoq\, O.\, Neglia\, G.\, Vidal\
 , R.\, & Kameni\, L. (2022). Personalized Federated Learning through Local
  Memorization. In Proceedings of the 39th International Conference on Mach
 ine Learning (PMLR 162:15070-15092). Available at:https://proceedings.mlr.
 press/v162/marfoq22a.html.\n- Marfoq\, O.\, Neglia\, G.\, Bellet\, A.\, Ka
 meni\, L.\, & Vidal\, R. (2021). Federated Multi-Task Learning under a Mix
 ture of Distributions. Advances in Neural Information Processing Systems 3
 4 (NeurIPS 2021). Available at:https://proceedings.neurips.cc/paper/2021/h
 ash/82599a4ec94aca066873c99b4c741ed8-Abstract.html.\n- Galante\, F.\, Negl
 ia\, G.\, & Leonardi\, E. (2024). Scalable Decentralized Algorithms for On
 line Personalized Mean Estimation. Soon to be available on arxiv.\n\nBio:\
 nDr. Giovanni Neglia is a researcher at Inria\, France\, since 2008\, and 
 holds a chair on Pervasive Sustainable Learning Systems at the 3IA Côte d
 'Azur (one of the French Interdisciplinary Institutes on Artificial Intell
 igence)\, since 2021. He received his Habilitation in 2017 from the Univer
 sité Côte d'Azur\, France\, his PhD and electronic engineering degree fr
 om University of Palermo\, Italy\, respectively in 2005 and in 2001. Befor
 e joining Inria as a permanent researcher\, he was a research scholar at t
 he University Massachusetts Amherst\, (2005) and a postdoc at Inria (2006-
 2007). His research activity focuses on modelling and performance evaluati
 on of networked systems and proposals of new mechanisms to improve their p
 erformance. Currently\, his main interests are federated learning and onli
 ne learning
LOCATION:ME C2 405 https://plan.epfl.ch/?room==ME%20C2%20405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR