Abstract
We consider active data selection and test point rejection strategies for Gaussian process regression based on the variance of the posterior over target values. Gaussian process regression is viewed as transductive regression that provides target distributions for given points rather than selecting an explicit regression function. Since not only the posterior mean but also the posterior variance are easily calculated we use this additional information to two ends: Active data selection is performed by either querying at points of high estimated posterior variance or at points that minimize the estimated posterior variance averaged over the input distribution of interest or — in a transductive manner — averaged over the test set. Test point rejection is performed using the estimated posterior variance as a confidence measure. We find for both a two-dimensional toy problem and for a real-world benchmark problem that the variance is a reasonable criterion for both active data selection and test point rejection.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Cohn, D.A., Neural networks exploration using optimal experiment design. Neural Networks 6(9), 1996.
Gibbs, M.N., Bayesian Gaussian processes for regression and Classification. PhD thesis, Cambridge University, 1997.
Graepel,T., Herbrich R., and Obermayer, K., Bayesian Transduction. Advances in Neural Information Processing Systems 11, MIT Press, 1999.
Mackay, D.J.C., Information-based objective functions for active data selection, Neural Computation, 4(4)589–603, 1992.
Mackay, D.J.C., Gaussian processes, Tutorial at NIPS 10, 1998.
Vapnik, V., Statistical Learning Theory. New York, John Wiley and Sons, 1998.
Rasmussen, C.E. and Williams, C.K.I. Gaussian processes for regression Advances in Neural Information Processing Systems 8, MIT press, 1996.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Seo, S., Wallat, M., Graepel, T., Obermayer, K. (2000). Gaussian Process Regression: Active Data Selection and Test Point Rejection. In: Sommer, G., Krüger, N., Perwass, C. (eds) Mustererkennung 2000. Informatik aktuell. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-59802-9_4
Download citation
DOI: https://doi.org/10.1007/978-3-642-59802-9_4
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-67886-1
Online ISBN: 978-3-642-59802-9
eBook Packages: Springer Book Archive