A supervised strategy for deep kernel machine
Florian Yger, Maxime Berar, Gilles Gasso and Alain Rakotomamonjy
In: ESANN 2011, 27 - 29 April 2011, Bruges, Belgium.
This paper presents an alternative to the supervised KPCA
based approach for learning a Multilayer Kernel Machine (MKM) . In
our proposed procedure, the hidden layers are learnt in a supervised fash-
ion based on kernel partial least squares regression. The main interest
resides in a simplified learning scheme as the obtained hidden features are
automatically ranked according to their correlation with the target out-
puts. The approach is illustrated on small scale real world applications
and shows compelling evidences.