Learning Hidden Markov Models using Non-Negative Matrix Factorization
Machine Learning
2011-01-11 v2 Artificial Intelligence
Information Theory
math.IT
Abstract
The Baum-Welsh algorithm together with its derivatives and variations has been the main technique for learning Hidden Markov Models (HMM) from observational data. We present an HMM learning algorithm based on the non-negative matrix factorization (NMF) of higher order Markovian statistics that is structurally different from the Baum-Welsh and its associated approaches. The described algorithm supports estimation of the number of recurrent states of an HMM and iterates the non-negative matrix factorization (NMF) algorithm to improve the learned HMM parameters. Numerical examples are provided as well.
Cite
@article{arxiv.0809.4086,
title = {Learning Hidden Markov Models using Non-Negative Matrix Factorization},
author = {George Cybenko and Valentino Crespi},
journal= {arXiv preprint arXiv:0809.4086},
year = {2011}
}
Comments
Submitted to IEEE Transactions on Information Theory in September 2008