BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Conductivity Imaging using Deep Neural Networks - Bangti Jin (Univ
 ersity College London)
DTSTART:20230519T103000Z
DTEND:20230519T113000Z
UID:TALK198145@talks.cam.ac.uk
DESCRIPTION:Conductivity imaging from various observational data represent
 s one fundamental task in medical imaging. In this talk\, we discuss numer
 ical methods for identifying the conductivity parameters in elliptic PDEs.
  Commonly\, a regularized formulation consists of a data fidelity and a re
 gularizer is employed\, and then it is discretized using finite difference
  method\, finite element methods or deep neural networks in practical comp
 utation. One key issue is to establish a priori error estimates for the re
 covered conductivity distribution. In this talk\, we discuss our recent fi
 ndings on using deep neural networks for this class of problems\, by effec
 tively utilizing relevant stability results.&nbsp\;
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR