BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Client Clustering for Federated Learning in Data Heterogeneous Sce
 narios - Gabriel Ukstin Talasso\, Campinas (Unicamp)\, Brazil
DTSTART:20251022T100000Z
DTEND:20251022T110000Z
UID:TALK238681@talks.cam.ac.uk
CONTACT:Sally Matthews
DESCRIPTION:Federated Learning (FL) is a paradigm where models are collabo
 ratively trained by sharing only local parameters with a central aggregati
 on server and faces limitations in heterogeneous environments. In particul
 ar\, the heterogeneity of client data and device capabilities affects mode
 l generalization\, convergence\, and resource management. In this scenario
 \, “client clustering” has emerged as a strategy to mitigate these iss
 ues\, enabling more efficient model aggregation\, improving convergence\, 
 and enhancing personalization across diverse data distributions.\n\nBio: G
 abriel Ukstin Talasso holds a Bachelor's degree in Statistics and is curre
 ntly a Master’s student in Computer Science at the University of Campina
 s (Unicamp)\, Brazil. His research focuses on training and fine-tuning lan
 guage models in distributed environments using federated learning\, partic
 ularly in scenarios with heterogeneous data.\n
LOCATION:Computer Lab\, SS03
END:VEVENT
END:VCALENDAR