ROBUST STATISTICS:
  Integration of parametric robust methods and non-parametric kernel based methods.

 

Financing: Research Foundation - Flanders (FWO)

Project reference Nr.: G.0499.04
Start: 2004-01-01
End: 2007-12-31

Description:
The aim of the project is the development of robust methods for kernel based techniques and support vector machines, to achieve an efficient development and integration of both research domains. The following research themes are included: robustification of kernel methods, theoretical analysis of robustness properties, robust model selection, integration of software, application on concrete cases in bioinformatics, chemometrie, process industry and time series analysis. The interaction between the involved research groups and the interaction between theoretical, algorithmic and application aspects are also important objectives.
 

SMC people involved in the project: