BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Learning Solution Operators for PDEs with Uncertainty
DTSTART:20240610T160000
DTEND:20240610T170000
DTSTAMP:20260610T173714Z
UID:e01fea5218e52be66d137997d6ffb97523618a8f1421e61bcb67e797
CATEGORIES:Conferences - Seminars
DESCRIPTION:Emilia Magnani\, ELLIS Program and University of Tübingen\nT
 he EPFL SIAM Student chapter is organizing a series of online seminars on 
 applied math.\n\nNext Monday\, June 10th\, at 16.00\, we will host Emilia
  Magnani from the ELLIS program and University of Tübingen.\n\n \n\nTitl
 e: Learning Solution Operators for PDEs with Uncertainty\n\n \n\nAbstract
 : We provide a Bayesian formulation of the problem of learning solution op
 erators of PDEs in the formalism of Gaussian processes. We consider neural
  operators\, recent deep architectures that have shown promising results i
 n tackling the task of learning PDE solution operators. The current state 
 of the art for these models lacks explicit uncertainty quantification. Our
  approach offers a practical and theoretically sound way to apply the line
 arized Laplace approximation to neural operators to provide uncertainty es
 timates. Moreover\, we introduce a new framework for Bayesian uncertainty 
 quantification in neural operators using function-valued Gaussian processe
 s.\n\n \n\nBio: Emilia Magnani is a Ph.D. candidate at the University of
  Tübingen under the supervision of Philipp Hennig. She is also part of t
 he ELLIS program and spent part of her Ph.D. in Genoa working with her co
 -supervisor\, Lorenzo Rosasco. Before that\, she attained a Master's degre
 e in Mathematics from ETH Zurich. Her research interests span various are
 as of machine learning such as probabilistic numerics\, Gaussian processes
 \, and operator learning.
LOCATION:https://epfl.zoom.us/j/63925499984?pwd=GOEI1rAQrMOXaIgFLG5B3IYle4
 Funr.1
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR