inria-00321505, version 1
The generative self-organizing map: a probabilistic generalization of Kohonen's SOM
Jakob Verbeek
1Nikos Vlassis a, 1Ben Krose b, 1
N° IAS-UVA-02-03 (2002)
Abstract: We present a variational Expectation-Maximization algorithm to learn proba- bilistic mixture models. The algorithm is similar to Kohonen's Self-Organizing Map algorithm and can be applied on any mixture model for which we can find a standard Expectation Maximization algorithm. We maximize the variational free- energy which sums data log-likelihood and Kullback-Leibler divergence between the neighborhood function and the posterior distribution on the components, given data. We illustrate the algorithm with an application on word clustering.
- a – Technical University of Crete
- b – University of Amsterdam
- 1: Instituut voor Informatica (IvI)
- Universiteit van Amsterdam
- Domain : Computer Science/Learning
- Keywords : self-organizing map – Gaussian mixture – free-energy minimization – Expectation-Maximization algorithm – vector quantization
- Internal note : IAS-UVA-02-03
- Comment : University of Amsterdam
- inria-00321505, version 1
- http://hal.inria.fr/inria-00321505
- oai:hal.inria.fr:inria-00321505
- From: Jakob Verbeek
- Submitted on: Wednesday, 16 February 2011 17:09:33
- Updated on: Friday, 18 February 2011 14:07:41
Associated documents
Export