BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Choosing a noise model for Electrical Impedance Tomography - Kweku
  Abraham\, University of Cambridge
DTSTART:20190206T160000Z
DTEND:20190206T170000Z
UID:TALK119563@talks.cam.ac.uk
CONTACT:Angeliki Menegaki
DESCRIPTION:One medical imaging technique is electrical impedance tomograp
 hy\, involving measuring voltages corresponding to different current profi
 les to infer conductivity. The associated inverse problem\, the Calderon p
 roblem\, has been widely studied. In inverse problems literature\, the mea
 surement error is assumed to be deterministic and small. For statisticians
 \, a more satisfactory approach allows large noise according to a specifie
 d noise model. But what should the noise model be in this case? A natural 
 candidate would be white noise in a Hilbert space\, but the appropriate Hi
 lbert space to use is not obvious. I will walk through the steps required 
 to reach such a Hilbert space.
LOCATION:MR14\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR