Products of Hidden Markov Models

Abstract

We present products of hidden Markov models (PoHMM's), a way of combining HMM's to form a distributed state time series model. Inference in a PoHMM is tractable and eÆcient. Learning of the parameters, although intractable, can be e ectively done using the Product of Experts learning rule. The distributed state helps the model to explain data which has… (More)

Topics

9 Figures and Tables

Cite this paper

@inproceedings{Brown2001ProductsOH, title={Products of Hidden Markov Models}, author={Andrew D. Brown and Geoffrey E. Hinton}, booktitle={AISTATS}, year={2001} }