BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Layered\, error enhanced hierarchical dictionary learning algorith
 m for sparse coding. - Daniela Calvetti (Case Western Reserve University)
DTSTART:20230428T103000Z
DTEND:20230428T113000Z
UID:TALK198454@talks.cam.ac.uk
DESCRIPTION:In this talk we preset a novel multi-phase dictionary learning
  algorithm that addresses the complexity by clustering and reducing the di
 ctionary\, enhancing the resolution power of the method by accounting for 
 the representation error introduced by the dictionary reduction. The probl
 em is set up and solved in the Bayesian framework\, and all steps involvin
 g sparse coding are performed by using sparsity promoting Bayesian hypermo
 dels and a priorconditioning techniques that are demonstrated earlier to p
 rovide a computationally efficient way to find compressible solutions to l
 inear inverse problems. As a novelty\, in the cluster identification probl
 em\, we introduce a new and data-informed way to implement group sparsity 
 in order to identify as few clusters as possible to explain the data. More
 over\, ideas from the previous works on Bayesian modeling error analysis a
 re modefied and extended to quantify the modeling error introduced when pa
 ssing from the full dictionary cluster to the reduced one.&nbsp\;\nThis wo
 rks has been done in collaboration with Alberto Bocchinfuso and Erkki Some
 rsalo.\n&nbsp\;
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR