Header logo is

Gaussian Processes for Regression

1996

Conference Paper

ei


The Bayesian analysis of neural networks is difficult because a simple prior over weights implies a complex prior over functions. We investigate the use of a Gaussian process prior over functions, which permits the predictive Bayesian analysis for fixed values of hyperparameters to be carried out exactly using matrix operations. Two methods, using optimization and averaging (via Hybrid Monte Carlo) over hyperparameters have been tested on a number of challenging problems and have produced excellent results.

Author(s): Williams, CKI. and Rasmussen, CE.
Book Title: Advances in neural information processing systems 8
Journal: Advances in Neural Processing Systems 8
Pages: 514-520
Year: 1996
Month: June
Day: 0
Editors: Touretzky, D.S. , M.C. Mozer, M.E. Hasselmo
Publisher: MIT Press

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: Ninth Annual Conference on Neural Information Processing Systems (NIPS 1995)
Event Place: Denver, CO, USA

Address: Cambridge, MA, USA
Digital: 0
ISBN: 0-262-20107-0
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{2468,
  title = {Gaussian Processes for Regression},
  author = {Williams, CKI. and Rasmussen, CE.},
  journal = {Advances in Neural Processing Systems 8},
  booktitle = {Advances in neural information processing systems 8},
  pages = {514-520},
  editors = {Touretzky, D.S. , M.C. Mozer, M.E. Hasselmo},
  publisher = {MIT Press},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Cambridge, MA, USA},
  month = jun,
  year = {1996},
  month_numeric = {6}
}