Paper available on combining estimators and Gaussian process regression

[Volker Tresp]

I would like to announce the availability of a new paper of potential
interest to people working on  combining estimators and Gaussian
process regression or other kernel based systems.

Comments are welcome!

Greetings,

- Volker

-------------------------------------------------------------------------

                     A BAYESIAN COMMITTEE MACHINE

                             Volker Tresp
                    Siemens AG, Corporate Technology
                          Neural Computation 
                    Dept. Information and Communications
                  Otto-Hahn-Ring 6, 81730 Munich, Germany 


The Bayesian committee machine (BCM) is a novel approach to combining
estimators which were trained on different data sets.  Although the BCM
can be applied to the combination of any kind of estimators the main
foci are Gaussian process regression and related systems such as
regularization networks and smoothing splines  for which the degrees of
freedom  increase with the number of training data.  Somewhat
surprisingly, we find that the performance of the BCM improves if
several test points are queried at the same time and is optimal if the
number of test points is at least as large as the degrees of freedom of
the estimator.  The BCM also provides a new solution  for online
learning with potential  applications to  data mining.  We apply the
BCM to systems with fixed basis functions and discuss its relationship
to Gaussian process regression. Finally, we also show how the ideas
behind the BCM can be applied in a frequentist setting to extend the
input dependent combination of estimators.


ftp://flop.informatik.tu-muenchen.de/pub/hofmannr/bcm.ps.gz
-----------------------------------------------------------------------
-----------------------------------------------------------------------