BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Learned Regularizers for Image Reconstruction
DTSTART:20260423T093000Z
DTEND:20260423T110000Z
DTSTAMP:20260428T220600Z
UID:indico-event-18767@indico.psi.ch
CONTACT:benjamin.bejar@psi.ch
DESCRIPTION:Speakers: Stanislas Ducotterd (EPFL)\n\nAbstract: This present
 ation explores the evolution of image reconstruction\, beginning with an i
 ntuitive overview of classical regularizers and their role in solving inve
 rse problems. It then examines modern deep-learning-driven iterative techn
 iques\, which\, despite their powerful denoising capabilities\, can strugg
 le with instability and fail to converge. Finally\, the talk surveys the c
 urrent state of research dedicated to overcoming this challenge\, illustra
 ting how the field is working to restore rigorous mathematical convergence
  guarantees without sacrificing the empirical performance of state-of-the-
 art neural networks.\nTEAMS link\n\nhttps://indico.psi.ch/event/18767/
LOCATION:OHSA/B17
URL:https://indico.psi.ch/event/18767/
END:VEVENT
END:VCALENDAR