Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Conference paper
Title: Hidden Markov model identifiability via tensors
Author: Tune, P.
Nguyen, H.
Roughan, M.
Citation: IEEE International Symposium on Information Theory, ISIT 2013, 2013/ pp.2299-2303
Publisher: IEEE
Publisher Place: Online
Issue Date: 2013
Series/Report no.: IEEE International Symposium on Information Theory
ISBN: 9781479904464
ISSN: 2157-8095
Conference Name: IEEE International Symposium on Information Theory (2013 : Istanbul)
Statement of
Paul Tune, Hung X. Nguyen and Matthew Roughan
Abstract: The prevalence of hidden Markov models (HMMs) in various applications of statistical signal processing and communications is a testament to the power and flexibility of the model. In this paper, we link the identifiability problem with tensor decomposition, in particular, the Canonical Polyadic decomposition. Using recent results in deriving uniqueness conditions for tensor decomposition, we are able to provide a necessary and sufficient condition for the identification of the parameters of discrete time finite alphabet HMMs. This result resolves a long standing open problem regarding the derivation of a necessary and sufficient condition for uniquely identifying an HMM. We then further extend recent preliminary work on the identification of HMMs with multiple observers by deriving necessary and sufficient conditions for identifiability in this setting.
Rights: ©2013 IEEE
RMID: 0020134746
DOI: 10.1109/ISIT.2013.6620636
Published version:
Appears in Collections:Mathematical Sciences publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.