Incremental Gaussian Processes

Quinonero Candela, J; Becker,; S.,; Thrun, S.; Obermayer, K.

Item

ITEM ACTIONSEXPORT

Add to Basket

Please note that a newer version of this item is available:
https://pure.mpg.de/pubman/item/item_1792290_2

DetailsSummary

Released

Conference Paper

Incremental Gaussian Processes

MPS-Authors

/persons/resource/persons83845

Quinonero Candela, J
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

External Resource

No external resources are shared

Fulltext (restricted access)

There are currently no full texts shared for your IP range.

Fulltext (public)

There are no public fulltexts stored in PuRe

Supplementary Material (public)

There is no public supplementary material available

Citation

Quinonero Candela, J. (2003). Incremental Gaussian Processes. Advances in Neural Information Processing Systems 15, 1001-1008.

Cite as: https://hdl.handle.net/11858/00-001M-0000-0013-DB41-7

Abstract

In this paper, we consider Tipping‘s relevance vector machine (RVM) and formalize an incremental training strategy as a variant of the expectation-maximization (EM) algorithm that we call subspace EM. Working with a subset of active basis functions, the sparsity of the RVM solution will ensure that the number of basis functions and thereby the computational complexity is kept low. We also introduce a mean field approach to the intractable classification model that is expected to give a very good approximation to exact Bayesian inference and contains the Laplace approximation as a special case. We test the algorithms on two large data sets with O(10^3-10^4) examples. The results indicate that Bayesian learning of large data sets, e.g. the MNIST database is realistic.