Sparse and Spurious: Dictionary Learning with Noise and Outliers
Event details
| Date | 23.02.2015 |
| Hour | 16:15 |
| Speaker |
Rémi Gribonval / PANAMA / INRIA Rennes Bio: Rémi Gribonval is a Research Director (Directeur de Recherche) with INRIA in Rennes, France, and the scientific leader of the PANAMA research group on sparse audio processing. In 2011, he was awarded the Blaise Pascal Award of the GAMNI-SMAI by the French Academy of Sciences, and a starting investigator grant from the European Research Council in 2011. He is an IEEE fellow. He founded the series of international workshops SPARS on Signal Processing with Adaptive/Sparse Representations. Since 2002 he has been the coordinator of several national, bilateral and European research projects. He is currently a member of the LVA/ICA steering committee, the IEEE SPTM Technical Committee, and the SPARS steering committee. Rémi Gribonval was a student at Ecole Normale Supérieure, Paris from 1993 to 1997. He received the Ph. D. degree in applied mathematics from the University of Paris-IX Dauphine, Paris, France, in 1999, and his Habilitation à Diriger des Recherches in applied mathematics from the University of Rennes I, Rennes, France, in 2007. |
| Location | |
| Category | Conferences - Seminars |
Many tasks, ranging from the resolution of inverse problems to denoising, can be efficiently addressed assuming some sparse model with an overcomplete dictionary. In the last decade, the statistical and algorithmic analysis of these approaches has become quite mature. Yet, the choice of the dictionary for a given problem remains a key practical issue, empirically addressed through data-driven principles known as dictionary learning.
The talk will first briefly review dictionary learning and related sparse matrix factorizations, then I will describe recently obtained generalization bounds and identifiability guarantees for dictionary learning in the presence of noise and outliers.
The talk will first briefly review dictionary learning and related sparse matrix factorizations, then I will describe recently obtained generalization bounds and identifiability guarantees for dictionary learning in the presence of noise and outliers.
Practical information
- General public
- Free
Organizer
- LCAV
Contact
- Ivan Dokmanic