Dimension reduction for hidden Markov models using the suficiency approach

Dimension reduction is often included in pattern recognizers based on hidden Markov models to lower the size of the models to estimate. Commonly used methods are heuristic in nature and do not take care of information retention after projection. In this paper, we present a new method based on the ap...

Descripción completa

Detalles Bibliográficos
Autores principales: Tomassi, Diego, Forzani, Liliana, Milone, Diego Humberto, Cook, R. Dennis
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2011
Materias:
Ciencias Informáticas
Markov models
Dimension reduction
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/125252
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-125252
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Markov models
Dimension reduction
spellingShingle Ciencias Informáticas
Markov models
Dimension reduction
Tomassi, Diego
Forzani, Liliana
Milone, Diego Humberto
Cook, R. Dennis
Dimension reduction for hidden Markov models using the suficiency approach
topic_facet Ciencias Informáticas
Markov models
Dimension reduction
description Dimension reduction is often included in pattern recognizers based on hidden Markov models to lower the size of the models to estimate. Commonly used methods are heuristic in nature and do not take care of information retention after projection. In this paper, we present a new method based on the approach of suficient dimension reductions. It explicitly accounts for all the discriminative information available in the original features, while using a minimum number of linear combinations of them. We review the underlying theory and present an algorithm for the practical implementation of the proposed method. On the experimental side, we use simulations to illustrate its advantages over widely-used existing alternatives. In particular, we show that it performs as good as existing techniques when data is optimal according to the assumptions of those techniques, but signi cantly better for heteroscedastic data with no special structure on the covariance matrix.
format Objeto de conferencia
Objeto de conferencia
author Tomassi, Diego
Forzani, Liliana
Milone, Diego Humberto
Cook, R. Dennis
author_facet Tomassi, Diego
Forzani, Liliana
Milone, Diego Humberto
Cook, R. Dennis
author_sort Tomassi, Diego
title Dimension reduction for hidden Markov models using the suficiency approach
title_short Dimension reduction for hidden Markov models using the suficiency approach
title_full Dimension reduction for hidden Markov models using the suficiency approach
title_fullStr Dimension reduction for hidden Markov models using the suficiency approach
title_full_unstemmed Dimension reduction for hidden Markov models using the suficiency approach
title_sort dimension reduction for hidden markov models using the suficiency approach
publishDate 2011
url http://sedici.unlp.edu.ar/handle/10915/125252
work_keys_str_mv AT tomassidiego dimensionreductionforhiddenmarkovmodelsusingthesuficiencyapproach
AT forzanililiana dimensionreductionforhiddenmarkovmodelsusingthesuficiencyapproach
AT milonediegohumberto dimensionreductionforhiddenmarkovmodelsusingthesuficiencyapproach
AT cookrdennis dimensionreductionforhiddenmarkovmodelsusingthesuficiencyapproach
bdutipo_str Repositorios
_version_ 1764820451455926273

Ejemplares similares