BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning from mistakes - learning optimally sparse image filters b
 y quotient minimisation - Carola-Bibiane Schönlieb (University of Cambrid
 ge)
DTSTART:20180117T114500Z
DTEND:20180117T123000Z
UID:TALK97726@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span>Co-authors: Martin Benning		(University of Cambridge)\, 
 Guy Gilboa		(Technion\, Haifa)\, Joana Grah		(University of Cambridge)    
     <br></span><span><br>Learning approaches have recently become very pop
 ular in the field of inverse problems. A large variety of methods has been
  established in recent years\, ranging from bi-level learning to high-dime
 nsional machine learning techniques. Most learning approaches\, however\, 
 only aim at fitting parametrised models to favourable training data whilst
  ig- noring misfit training data completely. In this talk\, we fol- low up
  on the idea of learning parametrised regularisation functions by quotient
  minimisation. We consider one- and higher-dimensional filter functions to
  be learned and allow for fit- and misfit-training data consisting of mult
 iple func- tions. We first present results resembling behaviour of well- e
 stablished derivative-based sparse regularisers like total variation or hi
 gher-order total variation in one-dimension. Then\, we introduce novel fam
 ilies of non-derivative-based regularisers. This is accomplished by learni
 ng favourable scales and geometric properties while at the same time avoid
 ing unfavourable ones.</span>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR