Entropy-Isomap: Manifold Learning for High-dimensional Dynamic Processes
Scientific and engineering processes deliver massive high-dimensional data sets that are generated as non-linear transformations of an initial state and few process parameters. Mapping such data to a low-dimensional manifold facilitates better understanding of the underlying processes, and enables t...
Gespeichert in:
Hauptverfasser: | , , , , |
---|---|
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 | Schoeneman, Frank Chandola, Varun Napp, Nils Wodo, Olga Zola, Jaroslaw |
description | Scientific and engineering processes deliver massive high-dimensional data
sets that are generated as non-linear transformations of an initial state and
few process parameters. Mapping such data to a low-dimensional manifold
facilitates better understanding of the underlying processes, and enables their
optimization. In this paper, we first show that off-the-shelf non-linear
spectral dimensionality reduction methods, e.g., Isomap, fail for such data,
primarily due to the presence of strong temporal correlations. Then, we propose
a novel method, Entropy-Isomap, to address the issue. The proposed method is
successfully applied to large data describing a fabrication process of organic
materials. The resulting low-dimensional representation correctly captures
process control variables, allows for low-dimensional visualization of the
material morphology evolution, and provides key insights to improve the
process. |
doi_str_mv | 10.48550/arxiv.1802.06823 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1802_06823</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1802_06823</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-b1ead3ca2915cb3d9f045b2e122eafe7a2d0ad6cde24fd89aa9be0756857ff2f3</originalsourceid><addsrcrecordid>eNotz7FOwzAUQFEvDKjlA5jwDzg4dpw43VAppFIqGLpHL_ZzsZTYkY0Q-XtEYbrblQ4h9yUvKq0Uf4T07b-KUnNR8FoLeUu6Q_hMcVnZMccZlh09QfAuTpb2CCn4cKEuJtr5ywezfsaQfQww0ec1wOwNfU_RYM6Yt-TGwZTx7r8bcn45nPcd699ej_unnkHdSDaWCFYaEG2pzCht63ilRoGlEAgOGxCWg62NRVE5q1uAdkTeqFqrxjnh5IY8_G2vlGFJfoa0Dr-k4UqSPwRtSCA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Entropy-Isomap: Manifold Learning for High-dimensional Dynamic Processes</title><source>arXiv.org</source><creator>Schoeneman, Frank ; Chandola, Varun ; Napp, Nils ; Wodo, Olga ; Zola, Jaroslaw</creator><creatorcontrib>Schoeneman, Frank ; Chandola, Varun ; Napp, Nils ; Wodo, Olga ; Zola, Jaroslaw</creatorcontrib><description>Scientific and engineering processes deliver massive high-dimensional data
sets that are generated as non-linear transformations of an initial state and
few process parameters. Mapping such data to a low-dimensional manifold
facilitates better understanding of the underlying processes, and enables their
optimization. In this paper, we first show that off-the-shelf non-linear
spectral dimensionality reduction methods, e.g., Isomap, fail for such data,
primarily due to the presence of strong temporal correlations. Then, we propose
a novel method, Entropy-Isomap, to address the issue. The proposed method is
successfully applied to large data describing a fabrication process of organic
materials. The resulting low-dimensional representation correctly captures
process control variables, allows for low-dimensional visualization of the
material morphology evolution, and provides key insights to improve the
process.</description><identifier>DOI: 10.48550/arxiv.1802.06823</identifier><language>eng</language><subject>Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2018-02</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/1802.06823$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1802.06823$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Schoeneman, Frank</creatorcontrib><creatorcontrib>Chandola, Varun</creatorcontrib><creatorcontrib>Napp, Nils</creatorcontrib><creatorcontrib>Wodo, Olga</creatorcontrib><creatorcontrib>Zola, Jaroslaw</creatorcontrib><title>Entropy-Isomap: Manifold Learning for High-dimensional Dynamic Processes</title><description>Scientific and engineering processes deliver massive high-dimensional data
sets that are generated as non-linear transformations of an initial state and
few process parameters. Mapping such data to a low-dimensional manifold
facilitates better understanding of the underlying processes, and enables their
optimization. In this paper, we first show that off-the-shelf non-linear
spectral dimensionality reduction methods, e.g., Isomap, fail for such data,
primarily due to the presence of strong temporal correlations. Then, we propose
a novel method, Entropy-Isomap, to address the issue. The proposed method is
successfully applied to large data describing a fabrication process of organic
materials. The resulting low-dimensional representation correctly captures
process control variables, allows for low-dimensional visualization of the
material morphology evolution, and provides key insights to improve the
process.</description><subject>Computer Science - Learning</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz7FOwzAUQFEvDKjlA5jwDzg4dpw43VAppFIqGLpHL_ZzsZTYkY0Q-XtEYbrblQ4h9yUvKq0Uf4T07b-KUnNR8FoLeUu6Q_hMcVnZMccZlh09QfAuTpb2CCn4cKEuJtr5ywezfsaQfQww0ec1wOwNfU_RYM6Yt-TGwZTx7r8bcn45nPcd699ej_unnkHdSDaWCFYaEG2pzCht63ilRoGlEAgOGxCWg62NRVE5q1uAdkTeqFqrxjnh5IY8_G2vlGFJfoa0Dr-k4UqSPwRtSCA</recordid><startdate>20180219</startdate><enddate>20180219</enddate><creator>Schoeneman, Frank</creator><creator>Chandola, Varun</creator><creator>Napp, Nils</creator><creator>Wodo, Olga</creator><creator>Zola, Jaroslaw</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20180219</creationdate><title>Entropy-Isomap: Manifold Learning for High-dimensional Dynamic Processes</title><author>Schoeneman, Frank ; Chandola, Varun ; Napp, Nils ; Wodo, Olga ; Zola, Jaroslaw</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-b1ead3ca2915cb3d9f045b2e122eafe7a2d0ad6cde24fd89aa9be0756857ff2f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Schoeneman, Frank</creatorcontrib><creatorcontrib>Chandola, Varun</creatorcontrib><creatorcontrib>Napp, Nils</creatorcontrib><creatorcontrib>Wodo, Olga</creatorcontrib><creatorcontrib>Zola, Jaroslaw</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>Schoeneman, Frank</au><au>Chandola, Varun</au><au>Napp, Nils</au><au>Wodo, Olga</au><au>Zola, Jaroslaw</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Entropy-Isomap: Manifold Learning for High-dimensional Dynamic Processes</atitle><date>2018-02-19</date><risdate>2018</risdate><abstract>Scientific and engineering processes deliver massive high-dimensional data
sets that are generated as non-linear transformations of an initial state and
few process parameters. Mapping such data to a low-dimensional manifold
facilitates better understanding of the underlying processes, and enables their
optimization. In this paper, we first show that off-the-shelf non-linear
spectral dimensionality reduction methods, e.g., Isomap, fail for such data,
primarily due to the presence of strong temporal correlations. Then, we propose
a novel method, Entropy-Isomap, to address the issue. The proposed method is
successfully applied to large data describing a fabrication process of organic
materials. The resulting low-dimensional representation correctly captures
process control variables, allows for low-dimensional visualization of the
material morphology evolution, and provides key insights to improve the
process.</abstract><doi>10.48550/arxiv.1802.06823</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.1802.06823 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_1802_06823 |
source | arXiv.org |
subjects | Computer Science - Learning Statistics - Machine Learning |
title | Entropy-Isomap: Manifold Learning for High-dimensional Dynamic Processes |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T01%3A45%3A13IST&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=Entropy-Isomap:%20Manifold%20Learning%20for%20High-dimensional%20Dynamic%20Processes&rft.au=Schoeneman,%20Frank&rft.date=2018-02-19&rft_id=info:doi/10.48550/arxiv.1802.06823&rft_dat=%3Carxiv_GOX%3E1802_06823%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 |