Mattia Fedrigo
homepage
GSF ISS MedPhys
Gebäude 34
Ingolstädter Landstraße 1
D-85764 Neuherberg, Germany
E-mail: mattia.fedrigo@gsf.de
Research
Probability theory and Information theory are my core interests.
Currently I am working on the problem of denoising and reconstructing a tomographic image
by means of a transform based on orthogonal polynomials on the disk.
I am also interested in two problems of parsimonious data representation:
the problem of reconstructing a sparse vector (a vector whose entries
are in large part zero) given a noisy linear underdetermined observation,
and the problem of efficiently encoding an image's wedgelet description.
Here is a
presentation (PDF, 1.23MB)
of a wedgelet image compression algorithm I gave at the
HASSIP 06
workshop, and a link to the
IBB wedgelet
homepage.
Papers
A Large Deviation Principle for the free energy of random Gibbs measures with application to the REM
,
a joint work with F. Flandoli and F. Morandin published on the Annali di Matematica Pura ed Applicata
in 2006.
Shannon Coding Theorem: a rigorous proof by large deviations properties of a disordered system
(PDF, 282KB),
a monograph with F. Flandoli and F. Morandin.
Curriculum
Here is a copy of my curriculum vitae (RTF, 32KB) updated the 25th of May 2007.