BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Operator learning in Uncertainty Quantification
DTSTART;VALUE=DATE-TIME:20221206T160000
DTEND;VALUE=DATE-TIME:20221206T170000
UID:98e1cf3e58e7a37ebccdf59550efa6b7b1f3f1e4f7427f16ee1487ff
CATEGORIES:Conferences - Seminars
DESCRIPTION:Jakob Zech (Heidelberg)\nIn this talk\, we discuss operator le
arning for computing forward surrogates of data-to-solution maps that occu
r in UQ. It is assumed that the data-to-solution operator is a complex dif
ferentiable mapping between two separable Hilbert spaces and that its inpu
ts and outputs are parameterized by stable affine representation systems s
uch as orthonormal bases or frames. Dimension-independent algebraic bounds
on the expression rate with respect to the number of network parameters a
re established. We discuss possible applications such as parametric soluti
ons for second order elliptic PDEs in polygonal domains.
LOCATION:GA 3 21 https://plan.epfl.ch/?room==GA%203%2021
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR