From explained variance of correlated components to PCA without orthogonality constraints

Block Principal Component Analysis (Block PCA) of a data matrix A, where loadings Z are determined by maximization of AZ 2 over unit norm orthogonal loadings, is difficult to use for the design of sparse PCA by 1 regularization, due to the difficulty of taking care of both the orthogonality constrai...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chavent, Marie, Chavent, Guy
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 Chavent, Marie
Chavent, Guy
description Block Principal Component Analysis (Block PCA) of a data matrix A, where loadings Z are determined by maximization of AZ 2 over unit norm orthogonal loadings, is difficult to use for the design of sparse PCA by 1 regularization, due to the difficulty of taking care of both the orthogonality constraint on loadings and the non differentiable 1 penalty. Our objective in this paper is to relax the orthogonality constraint on loadings by introducing new objective functions expvar(Y) which measure the part of the variance of the data matrix A explained by correlated components Y = AZ. So we propose first a comprehensive study of mathematical and numerical properties of expvar(Y) for two existing definitions Zou et al. [2006], Shen and Huang [2008] and four new definitions. Then we show that only two of these explained variance are fit to use as objective function in block PCA formulations for A rid of orthogonality constraints.
doi_str_mv 10.48550/arxiv.2402.04692
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2402_04692</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2402_04692</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2402_046923</originalsourceid><addsrcrecordid>eNqFjjEKwkAQRbexEPUAVs4FjGtMREsJBksLG6swxI0ubHbC7BiT2xvF3urB5_F5Ss3XOkp2aapXyJ1tozjRcaST7T4eq2vOVIPpGofWmxu0yBZ9aYAqKInZOJRhLqluyBsvAYTgnB3gZeVBTwHigXfy6Kz0g-eD8HAlYapGFbpgZj9O1CI_XrLT8htRNGxr5L74xBTfmM1_4w3nCEER</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>From explained variance of correlated components to PCA without orthogonality constraints</title><source>arXiv.org</source><creator>Chavent, Marie ; Chavent, Guy</creator><creatorcontrib>Chavent, Marie ; Chavent, Guy</creatorcontrib><description>Block Principal Component Analysis (Block PCA) of a data matrix A, where loadings Z are determined by maximization of AZ 2 over unit norm orthogonal loadings, is difficult to use for the design of sparse PCA by 1 regularization, due to the difficulty of taking care of both the orthogonality constraint on loadings and the non differentiable 1 penalty. Our objective in this paper is to relax the orthogonality constraint on loadings by introducing new objective functions expvar(Y) which measure the part of the variance of the data matrix A explained by correlated components Y = AZ. So we propose first a comprehensive study of mathematical and numerical properties of expvar(Y) for two existing definitions Zou et al. [2006], Shen and Huang [2008] and four new definitions. Then we show that only two of these explained variance are fit to use as objective function in block PCA formulations for A rid of orthogonality constraints.</description><identifier>DOI: 10.48550/arxiv.2402.04692</identifier><language>eng</language><subject>Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2024-02</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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2402.04692$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2402.04692$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Chavent, Marie</creatorcontrib><creatorcontrib>Chavent, Guy</creatorcontrib><title>From explained variance of correlated components to PCA without orthogonality constraints</title><description>Block Principal Component Analysis (Block PCA) of a data matrix A, where loadings Z are determined by maximization of AZ 2 over unit norm orthogonal loadings, is difficult to use for the design of sparse PCA by 1 regularization, due to the difficulty of taking care of both the orthogonality constraint on loadings and the non differentiable 1 penalty. Our objective in this paper is to relax the orthogonality constraint on loadings by introducing new objective functions expvar(Y) which measure the part of the variance of the data matrix A explained by correlated components Y = AZ. So we propose first a comprehensive study of mathematical and numerical properties of expvar(Y) for two existing definitions Zou et al. [2006], Shen and Huang [2008] and four new definitions. Then we show that only two of these explained variance are fit to use as objective function in block PCA formulations for A rid of orthogonality constraints.</description><subject>Computer Science - Learning</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNqFjjEKwkAQRbexEPUAVs4FjGtMREsJBksLG6swxI0ubHbC7BiT2xvF3urB5_F5Ss3XOkp2aapXyJ1tozjRcaST7T4eq2vOVIPpGofWmxu0yBZ9aYAqKInZOJRhLqluyBsvAYTgnB3gZeVBTwHigXfy6Kz0g-eD8HAlYapGFbpgZj9O1CI_XrLT8htRNGxr5L74xBTfmM1_4w3nCEER</recordid><startdate>20240207</startdate><enddate>20240207</enddate><creator>Chavent, Marie</creator><creator>Chavent, Guy</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20240207</creationdate><title>From explained variance of correlated components to PCA without orthogonality constraints</title><author>Chavent, Marie ; Chavent, Guy</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2402_046923</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Chavent, Marie</creatorcontrib><creatorcontrib>Chavent, Guy</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>Chavent, Marie</au><au>Chavent, Guy</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>From explained variance of correlated components to PCA without orthogonality constraints</atitle><date>2024-02-07</date><risdate>2024</risdate><abstract>Block Principal Component Analysis (Block PCA) of a data matrix A, where loadings Z are determined by maximization of AZ 2 over unit norm orthogonal loadings, is difficult to use for the design of sparse PCA by 1 regularization, due to the difficulty of taking care of both the orthogonality constraint on loadings and the non differentiable 1 penalty. Our objective in this paper is to relax the orthogonality constraint on loadings by introducing new objective functions expvar(Y) which measure the part of the variance of the data matrix A explained by correlated components Y = AZ. So we propose first a comprehensive study of mathematical and numerical properties of expvar(Y) for two existing definitions Zou et al. [2006], Shen and Huang [2008] and four new definitions. Then we show that only two of these explained variance are fit to use as objective function in block PCA formulations for A rid of orthogonality constraints.</abstract><doi>10.48550/arxiv.2402.04692</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2402.04692
ispartof
issn
language eng
recordid cdi_arxiv_primary_2402_04692
source arXiv.org
subjects Computer Science - Learning
Statistics - Machine Learning
title From explained variance of correlated components to PCA without orthogonality constraints
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-16T00%3A44%3A07IST&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=From%20explained%20variance%20of%20correlated%20components%20to%20PCA%20without%20orthogonality%20constraints&rft.au=Chavent,%20Marie&rft.date=2024-02-07&rft_id=info:doi/10.48550/arxiv.2402.04692&rft_dat=%3Carxiv_GOX%3E2402_04692%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