SEMINARIOS DE INVESTIGACIÓN EN INGENIERÍA INFORMÁTICA Y DE TELECOMUNICACIÓN 2007-2008


Actividad de Formación Continua  del Programa Oficial de Posgrado en Ingeniería Informática y de Telecomunicación


Escuela Politécnica Superior, Universidad Autónoma de Madrid

Escuela Politécnica Superior                        

viernes, 19 de octubre de 2007, 11:00

Seminario B-351, Escuela Politécnica Superior, Universidad Autónoma de Madrid

Bayesian machine learning with applications to brains, genes, and hearing aids.

Dr. Tom Heskes

Department of Information and Knowledge Systems (IRIS)
Nijmegen Institute for Computing and Information Sciences
Radboud University Nijmegen
The Netherlands

     

Abstract

Machine learning is about learning models from data. In so-called Bayesian machine learning we build probabilistic models and use probability calculus, in particular Bayes' rule, to infer the unknown model parameters given the observed data. On the one hand, we aim to build richer models that better match the characteristics of real-world data and are also able to incorporate the available domain knowledge. On the other hand, even for the simplest models exact probabilistic inference is intractable. A lot of our effort then goes into the design, understanding, and evaluation of algorithms for approximate probabilistic inference.
In my presentation I will show where this leads to by highlighting some of the applications that we work on: brain-computer interfacing (how to control devices by reading out brain activity), functional genomics (how to use functional and structural data to unravel the life cycle of the malaria parasite), and personalization of hearing aids (how to design listening experiments that reveal the preferences of individual users).

PDF presentation

Tom Heskes CV

Tom Heskes is associate professor in computing and information science at the Radboud University Nijmegen. After his PhD in 1993, he worked as a postdoc at the Beckman Institute, Champaign-Urbana, Illinois. Back in the Netherlands he became a research associate at SNN, formerly the Dutch Foundation for Neural Networks. Here he also started and headed a spin-off company, commercializing applications of machine learning and probabilistic artificial intelligence. In 2004 he moved from SNN to the computing science department. He is editor-in-chief of the journal Neurocomputing.