Rechercher

sur ce site


Accueil du site > Résumés des séminaires > Labo > Spectral Methods for Learning from High Dimensional Data

Spectral Methods for Learning from High Dimensional Data

In this talk we discuss some recent results for learning from high dimensional data sets arising in a wide variety of applications. The tools we build upon are spectral and analytical methods that allow to model complex data, while achieving good computational and theoretical properties. Empirically the derived algorithms obtain state of the art performances both on simulated and real data. Interestingly, our approach highlights the connections between learning theory and other fields in applied sciences, such as statistics, signal processing and inverse problems.

CMAP UMR 7641 École Polytechnique CNRS, Route de Saclay, 91128 Palaiseau Cedex France, Tél: +33 1 69 33 46 23 Fax: +33 1 69 33 46 46