Colloquium Mathematics, Antonino Abbruzzo

 

Join us for coffee and tea at 15.45 p.m.

 

 

Date: Thursday, April 5^th 2012

Speaker: Antonino Abbruzzo

Room: 5161.0253 (Bernoulliborg)

Time: 16.15

 

 

 

 

Title: Graphical models for estimating dynamic networks

 

 

Abstract:

 

Graphical models are powerful tools to estimate complex networks such as genomic networks. A genomic network is a system of relationships between genes which are represented by nodes. These relationships correspond to conditional dependences between genes, and they are represented by links. If we do not have any information about the presence or absence of links, then we need to recover these links from the data. The complexity of estimating network structures result from the very large number of possible models, that is given by 2^C(p,2) where C(p,2) is the binomial coefficient and p is the number of nodes. Surprisingly, penalized graphical models can be applied to recover network structures even if p is large relatively to the number of observations. However, penalized graphical models do not consider the dynamic of the networks. We propose two new models, the factorial graphical lasso and the slowly changing graphical models, to estimate complex dynamic networks.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Colloquium coordinators are Prof.dr. A.C.D. van Enter (e-mail : A.C.D.van.Enter@rug.nl) and

Dr. A.V. Kiselev (e-mail: a.v.kiselev rug.nl )

 

http://www.rug.nl/wiskunde/nieuws/actueel/wiskundecolloquia