Distributionally Robust Graphical Models

In many structured prediction problems, complex relationships between variables are compactly defined using graphical structures. The most prevalent graphical prediction methods---probabilistic graphical models and large margin methods---have their own distinct strengths but also possess significant...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Fathony, Rizal, Rezaei, Ashkan, Bashiri, Mohammad Ali, Zhang, Xinhua, Ziebart, Brian D
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 Fathony, Rizal
Rezaei, Ashkan
Bashiri, Mohammad Ali
Zhang, Xinhua
Ziebart, Brian D
description In many structured prediction problems, complex relationships between variables are compactly defined using graphical structures. The most prevalent graphical prediction methods---probabilistic graphical models and large margin methods---have their own distinct strengths but also possess significant drawbacks. Conditional random fields (CRFs) are Fisher consistent, but they do not permit integration of customized loss metrics into their learning process. Large-margin models, such as structured support vector machines (SSVMs), have the flexibility to incorporate customized loss metrics, but lack Fisher consistency guarantees. We present adversarial graphical models (AGM), a distributionally robust approach for constructing a predictor that performs robustly for a class of data distributions defined using a graphical structure. Our approach enjoys both the flexibility of incorporating customized loss metrics into its design as well as the statistical guarantee of Fisher consistency. We present exact learning and prediction algorithms for AGM with time complexity similar to existing graphical models and show the practical benefits of our approach with experiments.
doi_str_mv 10.48550/arxiv.1811.02728
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1811_02728</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1811_02728</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-dcde8a97f156e27427729a99dd28ae282de7f1b342b35c3a4825c86844c3bfa43</originalsourceid><addsrcrecordid>eNotzj1vwjAUhWEvDBXwAzqRkSUhvrbjmxFRSiuBkBB7dP0RYckQ5IQK_n1b2ukMr3T0MPbKy0KiUuWC0j18FRw5L0rQgC9s_hb6IQVzG0J3oRgf2aEzt37INomup2ApZrvO-dhP2Kil2Pvp_47Z8X19XH3k2_3mc7Xc5lRpzJ11HqnWLVeVBy1Ba6iprp0DJA8Izv80IyQYoawgiaAsViilFaYlKcZs9nf7pDbXFM6UHs0vuXmSxTcnpztG</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Distributionally Robust Graphical Models</title><source>arXiv.org</source><creator>Fathony, Rizal ; Rezaei, Ashkan ; Bashiri, Mohammad Ali ; Zhang, Xinhua ; Ziebart, Brian D</creator><creatorcontrib>Fathony, Rizal ; Rezaei, Ashkan ; Bashiri, Mohammad Ali ; Zhang, Xinhua ; Ziebart, Brian D</creatorcontrib><description>In many structured prediction problems, complex relationships between variables are compactly defined using graphical structures. The most prevalent graphical prediction methods---probabilistic graphical models and large margin methods---have their own distinct strengths but also possess significant drawbacks. Conditional random fields (CRFs) are Fisher consistent, but they do not permit integration of customized loss metrics into their learning process. Large-margin models, such as structured support vector machines (SSVMs), have the flexibility to incorporate customized loss metrics, but lack Fisher consistency guarantees. We present adversarial graphical models (AGM), a distributionally robust approach for constructing a predictor that performs robustly for a class of data distributions defined using a graphical structure. Our approach enjoys both the flexibility of incorporating customized loss metrics into its design as well as the statistical guarantee of Fisher consistency. We present exact learning and prediction algorithms for AGM with time complexity similar to existing graphical models and show the practical benefits of our approach with experiments.</description><identifier>DOI: 10.48550/arxiv.1811.02728</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2018-11</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/1811.02728$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1811.02728$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Fathony, Rizal</creatorcontrib><creatorcontrib>Rezaei, Ashkan</creatorcontrib><creatorcontrib>Bashiri, Mohammad Ali</creatorcontrib><creatorcontrib>Zhang, Xinhua</creatorcontrib><creatorcontrib>Ziebart, Brian D</creatorcontrib><title>Distributionally Robust Graphical Models</title><description>In many structured prediction problems, complex relationships between variables are compactly defined using graphical structures. The most prevalent graphical prediction methods---probabilistic graphical models and large margin methods---have their own distinct strengths but also possess significant drawbacks. Conditional random fields (CRFs) are Fisher consistent, but they do not permit integration of customized loss metrics into their learning process. Large-margin models, such as structured support vector machines (SSVMs), have the flexibility to incorporate customized loss metrics, but lack Fisher consistency guarantees. We present adversarial graphical models (AGM), a distributionally robust approach for constructing a predictor that performs robustly for a class of data distributions defined using a graphical structure. Our approach enjoys both the flexibility of incorporating customized loss metrics into its design as well as the statistical guarantee of Fisher consistency. We present exact learning and prediction algorithms for AGM with time complexity similar to existing graphical models and show the practical benefits of our approach with experiments.</description><subject>Computer Science - Artificial Intelligence</subject><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>eNotzj1vwjAUhWEvDBXwAzqRkSUhvrbjmxFRSiuBkBB7dP0RYckQ5IQK_n1b2ukMr3T0MPbKy0KiUuWC0j18FRw5L0rQgC9s_hb6IQVzG0J3oRgf2aEzt37INomup2ApZrvO-dhP2Kil2Pvp_47Z8X19XH3k2_3mc7Xc5lRpzJ11HqnWLVeVBy1Ba6iprp0DJA8Izv80IyQYoawgiaAsViilFaYlKcZs9nf7pDbXFM6UHs0vuXmSxTcnpztG</recordid><startdate>20181106</startdate><enddate>20181106</enddate><creator>Fathony, Rizal</creator><creator>Rezaei, Ashkan</creator><creator>Bashiri, Mohammad Ali</creator><creator>Zhang, Xinhua</creator><creator>Ziebart, Brian D</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20181106</creationdate><title>Distributionally Robust Graphical Models</title><author>Fathony, Rizal ; Rezaei, Ashkan ; Bashiri, Mohammad Ali ; Zhang, Xinhua ; Ziebart, Brian D</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-dcde8a97f156e27427729a99dd28ae282de7f1b342b35c3a4825c86844c3bfa43</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Fathony, Rizal</creatorcontrib><creatorcontrib>Rezaei, Ashkan</creatorcontrib><creatorcontrib>Bashiri, Mohammad Ali</creatorcontrib><creatorcontrib>Zhang, Xinhua</creatorcontrib><creatorcontrib>Ziebart, Brian D</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>Fathony, Rizal</au><au>Rezaei, Ashkan</au><au>Bashiri, Mohammad Ali</au><au>Zhang, Xinhua</au><au>Ziebart, Brian D</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Distributionally Robust Graphical Models</atitle><date>2018-11-06</date><risdate>2018</risdate><abstract>In many structured prediction problems, complex relationships between variables are compactly defined using graphical structures. The most prevalent graphical prediction methods---probabilistic graphical models and large margin methods---have their own distinct strengths but also possess significant drawbacks. Conditional random fields (CRFs) are Fisher consistent, but they do not permit integration of customized loss metrics into their learning process. Large-margin models, such as structured support vector machines (SSVMs), have the flexibility to incorporate customized loss metrics, but lack Fisher consistency guarantees. We present adversarial graphical models (AGM), a distributionally robust approach for constructing a predictor that performs robustly for a class of data distributions defined using a graphical structure. Our approach enjoys both the flexibility of incorporating customized loss metrics into its design as well as the statistical guarantee of Fisher consistency. We present exact learning and prediction algorithms for AGM with time complexity similar to existing graphical models and show the practical benefits of our approach with experiments.</abstract><doi>10.48550/arxiv.1811.02728</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1811.02728
ispartof
issn
language eng
recordid cdi_arxiv_primary_1811_02728
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Learning
Statistics - Machine Learning
title Distributionally Robust Graphical Models
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-23T09%3A56%3A40IST&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=Distributionally%20Robust%20Graphical%20Models&rft.au=Fathony,%20Rizal&rft.date=2018-11-06&rft_id=info:doi/10.48550/arxiv.1811.02728&rft_dat=%3Carxiv_GOX%3E1811_02728%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