A Dictionary Learning Approach for Factorial Gaussian Models

In this paper, we develop a parameter estimation method for factorially parametrized models such as Factorial Gaussian Mixture Model and Factorial Hidden Markov Model. Our contributions are two-fold. First, we show that the emission matrix of the standard Factorial Model is unidentifiable even if th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Subakan, Y. Cem, Traa, Johannes, Smaragdis, Paris, Stein, Noah
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Subakan, Y. Cem
Traa, Johannes
Smaragdis, Paris
Stein, Noah
description In this paper, we develop a parameter estimation method for factorially parametrized models such as Factorial Gaussian Mixture Model and Factorial Hidden Markov Model. Our contributions are two-fold. First, we show that the emission matrix of the standard Factorial Model is unidentifiable even if the true assignment matrix is known. Secondly, we address the issue of identifiability by making a one component sharing assumption and derive a parameter learning algorithm for this case. Our approach is based on a dictionary learning problem of the form $X = O R$, where the goal is to learn the dictionary $O$ given the data matrix $X$. We argue that due to the specific structure of the activation matrix $R$ in the shared component factorial mixture model, and an incoherence assumption on the shared component, it is possible to extract the columns of the $O$ matrix without the need for alternating between the estimation of $O$ and $R$.
doi_str_mv 10.48550/arxiv.1508.04486
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1508_04486</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1508_04486</sourcerecordid><originalsourceid>FETCH-LOGICAL-a676-43e456837bf53577c3feb5b94aac8b1e4ef43c295cdc55fc3239c2c0542a777b3</originalsourceid><addsrcrecordid>eNotj7FOwzAUAL0woMIHMOEfSHDy_GJHYokKLUhBLN2j51ebWkrjyCkI_h4oTLed7oS4qVSpLaK6o_wZP8oKlS2V1ra5FPedfIh8immi_CV7T3mK05vs5jkn4oMMKcsN8SnlSKPc0vuyRJrkS9r7cbkSF4HGxV__cyV2m8fd-qnoX7fP664vqDFNocFrbCwYFxDQGIbgHbpWE7F1ldc-aOC6Rd4zYmCooeWaFeqajDEOVuL2T3vOH-Ycjz-xw-_GcN6Ab5LZQhk</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>A Dictionary Learning Approach for Factorial Gaussian Models</title><source>arXiv.org</source><creator>Subakan, Y. Cem ; Traa, Johannes ; Smaragdis, Paris ; Stein, Noah</creator><creatorcontrib>Subakan, Y. Cem ; Traa, Johannes ; Smaragdis, Paris ; Stein, Noah</creatorcontrib><description>In this paper, we develop a parameter estimation method for factorially parametrized models such as Factorial Gaussian Mixture Model and Factorial Hidden Markov Model. Our contributions are two-fold. First, we show that the emission matrix of the standard Factorial Model is unidentifiable even if the true assignment matrix is known. Secondly, we address the issue of identifiability by making a one component sharing assumption and derive a parameter learning algorithm for this case. Our approach is based on a dictionary learning problem of the form $X = O R$, where the goal is to learn the dictionary $O$ given the data matrix $X$. We argue that due to the specific structure of the activation matrix $R$ in the shared component factorial mixture model, and an incoherence assumption on the shared component, it is possible to extract the columns of the $O$ matrix without the need for alternating between the estimation of $O$ and $R$.</description><identifier>DOI: 10.48550/arxiv.1508.04486</identifier><language>eng</language><subject>Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2015-08</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1508.04486$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1508.04486$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Subakan, Y. Cem</creatorcontrib><creatorcontrib>Traa, Johannes</creatorcontrib><creatorcontrib>Smaragdis, Paris</creatorcontrib><creatorcontrib>Stein, Noah</creatorcontrib><title>A Dictionary Learning Approach for Factorial Gaussian Models</title><description>In this paper, we develop a parameter estimation method for factorially parametrized models such as Factorial Gaussian Mixture Model and Factorial Hidden Markov Model. Our contributions are two-fold. First, we show that the emission matrix of the standard Factorial Model is unidentifiable even if the true assignment matrix is known. Secondly, we address the issue of identifiability by making a one component sharing assumption and derive a parameter learning algorithm for this case. Our approach is based on a dictionary learning problem of the form $X = O R$, where the goal is to learn the dictionary $O$ given the data matrix $X$. We argue that due to the specific structure of the activation matrix $R$ in the shared component factorial mixture model, and an incoherence assumption on the shared component, it is possible to extract the columns of the $O$ matrix without the need for alternating between the estimation of $O$ and $R$.</description><subject>Computer Science - Learning</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2015</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj7FOwzAUAL0woMIHMOEfSHDy_GJHYokKLUhBLN2j51ebWkrjyCkI_h4oTLed7oS4qVSpLaK6o_wZP8oKlS2V1ra5FPedfIh8immi_CV7T3mK05vs5jkn4oMMKcsN8SnlSKPc0vuyRJrkS9r7cbkSF4HGxV__cyV2m8fd-qnoX7fP664vqDFNocFrbCwYFxDQGIbgHbpWE7F1ldc-aOC6Rd4zYmCooeWaFeqajDEOVuL2T3vOH-Ycjz-xw-_GcN6Ab5LZQhk</recordid><startdate>20150818</startdate><enddate>20150818</enddate><creator>Subakan, Y. Cem</creator><creator>Traa, Johannes</creator><creator>Smaragdis, Paris</creator><creator>Stein, Noah</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20150818</creationdate><title>A Dictionary Learning Approach for Factorial Gaussian Models</title><author>Subakan, Y. Cem ; Traa, Johannes ; Smaragdis, Paris ; Stein, Noah</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a676-43e456837bf53577c3feb5b94aac8b1e4ef43c295cdc55fc3239c2c0542a777b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2015</creationdate><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Subakan, Y. Cem</creatorcontrib><creatorcontrib>Traa, Johannes</creatorcontrib><creatorcontrib>Smaragdis, Paris</creatorcontrib><creatorcontrib>Stein, Noah</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Subakan, Y. Cem</au><au>Traa, Johannes</au><au>Smaragdis, Paris</au><au>Stein, Noah</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Dictionary Learning Approach for Factorial Gaussian Models</atitle><date>2015-08-18</date><risdate>2015</risdate><abstract>In this paper, we develop a parameter estimation method for factorially parametrized models such as Factorial Gaussian Mixture Model and Factorial Hidden Markov Model. Our contributions are two-fold. First, we show that the emission matrix of the standard Factorial Model is unidentifiable even if the true assignment matrix is known. Secondly, we address the issue of identifiability by making a one component sharing assumption and derive a parameter learning algorithm for this case. Our approach is based on a dictionary learning problem of the form $X = O R$, where the goal is to learn the dictionary $O$ given the data matrix $X$. We argue that due to the specific structure of the activation matrix $R$ in the shared component factorial mixture model, and an incoherence assumption on the shared component, it is possible to extract the columns of the $O$ matrix without the need for alternating between the estimation of $O$ and $R$.</abstract><doi>10.48550/arxiv.1508.04486</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1508.04486
ispartof
issn
language eng
recordid cdi_arxiv_primary_1508_04486
source arXiv.org
subjects Computer Science - Learning
Statistics - Machine Learning
title A Dictionary Learning Approach for Factorial Gaussian Models
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T16%3A17%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Dictionary%20Learning%20Approach%20for%20Factorial%20Gaussian%20Models&rft.au=Subakan,%20Y.%20Cem&rft.date=2015-08-18&rft_id=info:doi/10.48550/arxiv.1508.04486&rft_dat=%3Carxiv_GOX%3E1508_04486%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true