[Todos CMAT] Invitación Seminario IngeMat 2012 - Ignacio Ramírez - viernes 6 de julio - 9:00 hs.

[Marco Scavino - IMERL]

La Maestría en Ingeniería Matemática tiene el agrado de invitarles a
participar de la próxima ponencia 


del ciclo de Seminario IngeMat 2012, 

a cargo de Ignacio Ramírez 

quien presentará el trabajo titulado 
"An MDL framework for sparse coding and dictionary learning".
 
Horario y lugar: viernes 6 de julio, 9:00 horas, Salón Azul del Instituto de
Química, Facultad de Ingenieria. 
 
 ** Resumen **
 
The power of sparse signal modeling with learned over-complete
dictionaries has been demonstrated in a variety of applications and
fields, from signal processing to statistical inference and machine
learning. However, the statistical properties of these models, such as
under-fitting or over-fitting given sets of data, are still not well
characterized in the literature. As a result, the success of sparse
modeling depends on hand-tuning critical parameters for each data and
application. This work aims at addressing this by providing a
practical and objective characterization of sparse models by means of
the Minimum Description Length (MDL) principle -- a well established
information-theoretic approach to model selection in statistical
inference. The resulting framework derives a family of efficient
sparse coding and dictionary learning algorithms which, by virtue of
the MDL principle, are completely parameter free. Furthermore, such
framework allows to incorporate additional prior information to
existing models, such as Markovian dependencies, or to define
completely new problem formulations, including in the matrix analysis
area, in a natural way. These virtues will be demonstrated with
parameter-free algorithms for the classic image denoising and
classification problems, and for low-rank matrix recovery in video
applications.

El trabajo puede ser consultado en 
paper oficial en IEEE:   <http://dx.doi.org/10.1109/TSP.2012.2187203>
http://dx.doi.org/10.1109/TSP.2012.2187203
preprint arXiv:               <http://arxiv.org/abs/1110.2436>
http://arxiv.org/abs/1110.2436
 

Ignacio Ramírez
Procesamiento de Señales, Asistente (Gr.2)

Received the E.E. (2002), and the M.Sc. degree in Electrical
Engineering (2007) from the Universidad de la República, Uruguay
(UdelaR), and the Ph.D. degree in Scientific Computation (2012) from
the University of Minnesota (UofM).
He was a Research Assistant at the UofM from 2008 to 2012. He also
held temporary research positions at the UofM in 2007 and
Hewlett-Packard Laboratories, Palo Alto, in 2004.
Mr. Ramírez holds an Assistantship with the Department of
Electrical Engineering at UdelaR since 1999. His main research
interests are applied information theory statistical signal processing
abd machine learning, with focus in multimedia data processing. His
current research focuses in automatic model selection for sparse
linear models.

 
A partir de las 8:45 horas compartiremos café. 
 
SCAPA de Ingeniería Matemática



 
------------ próxima parte ------------
Se ha borrado un adjunto en formato HTML...
URL: <http://www.cmat.edu.uy/pipermail/todos/attachments/20120702/0d908e75/attachment.html>