BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:New ideas in optimal adaptivity: Time-dependent problems and Neura
 l networks [Hausdorff Colloquium]
DTSTART:20260610T130000Z
DTEND:20260610T143000Z
DTSTAMP:20260607T031100Z
UID:indico-event-1368@math-events.uni-bonn.de
DESCRIPTION:Speakers: Michael Feischl (TU Wien)\n\nAbstract:\nThe classica
 l theory of adaptive finite element methods is tailored to stationary PDEs
 . Recent developments made it possible to prove optimality results also fo
 r \nsimple time-dependent PDEs like the heat equation\, and also non-lin
 ear ODEs. These methods and ideas can be used to design and train neural n
 etworks\, that have an almost optimal\nnumber of parameters to reach a cer
 tain approximation error. The talk will give an overview over the approach
 \, present key steps in the proofs\, and show numerical examples.\nWebsite
  of the Hausdorff Colloquium\n\nhttps://math-events.uni-bonn.de/event/1368
 /
LOCATION:Endenicher Allee 60/1-016 - Lipschitzsaal (Mathezentrum)
URL:https://math-events.uni-bonn.de/event/1368/
END:VEVENT
END:VCALENDAR