Ensemble Learning of Coupled Parmeterised Kernel Models
Bart Hamers, Johan A. K. Suykens, Leemans, Bart De Moor · International Conference on Neural Information Processing · 2003
Abstract In this paper we propose a new method for learning a combination of estimators. Classically, committee networks are constructed after training the networks independently from each other. Here we present a learning strategy where the training is done in a coupled way. We illustrate that combining parameterized kernel methods with output coupling and use of a synchronization set of data points leads to an improved generalization. Examples are given on artificial and real life data sets.