Name: bcm
Owner: Open Data Science
Description: Bayesian Committee Machine MATLAB software written by Anton Schwaighofer
Created: 2015-05-28 15:21:00.0
Updated: 2016-01-18 22:19:43.0
Pushed: 2015-10-12 19:01:24.0
Size: 160
Language: Matlab
GitHub Committers
| User | Most Recent Commit | # Commits |
|---|
Other Committers
| User | Most Recent Commit | # Commits |
|---|
This software used to be hosted by Anton Schwaighofer, but doesn't appear to be on line any more. We've added it here under the sods repo to ensure it is available. Below is Anton's original README file.
Version 1.0, November 2005
The Bayesian Committee Machine (BCM) is an approximation method for large-scale Gaussian process regression.
What you should know beforehand:
The code is for Matlab
It requires the Netlab toolbox. You can download Netlab. Aston make netlab http://www.aston.ac.uk/eas/research/groups/ncrg/resources/netlab/ or we have a local repo available from sods https://github.com/sods/netlab/.
Install Netlab before trying to run any of the programs here.
To get started and to check your installation, try running 'dembcm.m'
If you are looking for example code to run the BCM, have a look at dembcm.m All of the main features are used and explained there.
Relevant publications:
V. Tresp. A bayesian committee machine. Neural Computation, 12, 2000
A. Schwaighofer and V. Tresp. Transductive and inductive methods for approximate Gaussian process regression. In S. Becker, S. Thrun, and K. Obermayer, editors, Advances in Neural Information Processing Systems