VIBES

Variational Inference for Bayesian Networks

SourceForge.net Logo
[Sourceforge project page]
John Winn, January 2004
Overview Tutorial Examples Help

Vibes Tutorial

This section contains a complete tutorial for using VIBES to perform variational inference in a Gaussian Mixture model. The model will then be modified to fit the data better (using the lower bound as an estimate of the log evidence for each model).

The tutorial consists of:
  1. Downloading the tutorial network and data files
  2. Loading the Matlab data into VIBES
  3. Creating and learning a Gaussian model
  4. Extending the Gaussian model to a Gaussian mixture model
  5. Inference in the Gaussian mixture model
  6. Modifying the mixture model
  7. Conclusion