Skip to main content
Kent Academic Repository

Estimating the Accuracy of Spectral Learning for HMMs

Liza, Farhana Ferdousi, Grzes, Marek (2016) Estimating the Accuracy of Spectral Learning for HMMs. In: Estimating the Accuracy of Spectral Learning for HMMs. Lecture Notes in Computer Science . pp. 46-56. Springer ISBN 978-3-319-44747-6. E-ISBN 978-3-319-44748-3. (doi:10.1007/978-3-319-44748-3_5) (KAR id:57317)

Abstract

Hidden Markov models (HMMs) are usually learned using

the expectation maximisation algorithm which is, unfortunately, subject

to local optima. Spectral learning for HMMs provides a unique, optimal

solution subject to availability of a sufficient amount of data. However,

with access to limited data, there is no means of estimating the accuracy

of the solution of a given model. In this paper, a new spectral evaluation

method has been proposed which can be used to assess whether the

algorithm is converging to a stable solution on a given dataset. The

proposed method is designed for real-life datasets where the true model is

not available. A number of empirical experiments on synthetic as well as

real datasets indicate that our criterion is an accurate proxy to measure

quality of models learned using spectral learning.

Item Type: Conference or workshop item (Paper)
DOI/Identification number: 10.1007/978-3-319-44748-3_5
Uncontrolled keywords: Spectral learning, HMM, SVD, Evaluation technique
Subjects: Q Science > Q Science (General) > Q335 Artificial intelligence
Q Science > QA Mathematics (inc Computing science) > QA273 Probabilities
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
Depositing User: Marek Grzes
Date Deposited: 15 Sep 2016 21:07 UTC
Last Modified: 09 Dec 2022 01:37 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/57317 (The current URI for this page, for reference purposes)

University of Kent Author Information

Liza, Farhana Ferdousi.

Creator's ORCID:
CReDIT Contributor Roles:

Grzes, Marek.

Creator's ORCID: https://orcid.org/0000-0003-4901-1539
CReDIT Contributor Roles:
  • Depositors only (login required):

Total unique views for this document in KAR since July 2020. For more details click on the image.