NFAD: Fixing anomaly detection using normalizing flows

Anomaly detection is a challenging task that frequently arises in practically all areas of industry and science, from fraud detection and data quality monitoring to finding rare cases of diseases and searching for new physics. Most of the conventional approaches to anomaly detection, such as one-cla...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2021-11
Hauptverfasser: Ryzhikov, Artem, Borisyak, Maxim, Ustyuzhanin, Andrey, Derkach, Denis
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Ryzhikov, Artem
Borisyak, Maxim
Ustyuzhanin, Andrey
Derkach, Denis
description Anomaly detection is a challenging task that frequently arises in practically all areas of industry and science, from fraud detection and data quality monitoring to finding rare cases of diseases and searching for new physics. Most of the conventional approaches to anomaly detection, such as one-class SVM and Robust Auto-Encoder, are one-class classification methods, i.e. focus on separating normal data from the rest of the space. Such methods are based on the assumption of separability of normal and anomalous classes, and subsequently do not take into account any available samples of anomalies. Nonetheless, in practical settings, some anomalous samples are often available; however, usually in amounts far lower than required for a balanced classification task, and the separability assumption might not always hold. This leads to an important task -- incorporating known anomalous samples into training procedures of anomaly detection models. In this work, we propose a novel model-agnostic training procedure to address this task. We reformulate one-class classification as a binary classification problem with normal data being distinguished from pseudo-anomalous samples. The pseudo-anomalous samples are drawn from low-density regions of a normalizing flow model by feeding tails of the latent distribution into the model. Such an approach allows to easily include known anomalies into the training process of an arbitrary classifier. We demonstrate that our approach shows comparable performance on one-class problems, and, most importantly, achieves comparable or superior results on tasks with variable amounts of known anomalies.
doi_str_mv 10.48550/arxiv.1912.09323
format Article
fullrecord <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_1912_09323</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2328980308</sourcerecordid><originalsourceid>FETCH-LOGICAL-a528-9ba663e23e8f185812d97dab7c7909dcc724e206b97c7b8adb3244ad2e17bba23</originalsourceid><addsrcrecordid>eNotj1FLwzAUhYMgOOZ-gE8WfG5Nbpom8W1MO4WhL3svSZNKRtfMpNXNX2-7-XQPH4fD_RC6IzjLBWP4UYWj-86IJJBhSYFeoRlQSlKRA9ygRYw7jDEUHBijM1S8l8vnp6R0R9d9Jqrze9WeEmN7W_fOd8kQJ975MHL3O-Wm9T_xFl03qo128X_naFu-bFev6eZj_bZablLFQKRSq6KgFqgVDRFMEDCSG6V5zSWWpq455BZwoeVItFBGU8hzZcASrrUCOkf3l9mzVHUIbq_CqZrkqrPc2Hi4NA7Bfw029tXOD6Ebf6qAgpACUyzoH-slUSY</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2328980308</pqid></control><display><type>article</type><title>NFAD: Fixing anomaly detection using normalizing flows</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Ryzhikov, Artem ; Borisyak, Maxim ; Ustyuzhanin, Andrey ; Derkach, Denis</creator><creatorcontrib>Ryzhikov, Artem ; Borisyak, Maxim ; Ustyuzhanin, Andrey ; Derkach, Denis</creatorcontrib><description>Anomaly detection is a challenging task that frequently arises in practically all areas of industry and science, from fraud detection and data quality monitoring to finding rare cases of diseases and searching for new physics. Most of the conventional approaches to anomaly detection, such as one-class SVM and Robust Auto-Encoder, are one-class classification methods, i.e. focus on separating normal data from the rest of the space. Such methods are based on the assumption of separability of normal and anomalous classes, and subsequently do not take into account any available samples of anomalies. Nonetheless, in practical settings, some anomalous samples are often available; however, usually in amounts far lower than required for a balanced classification task, and the separability assumption might not always hold. This leads to an important task -- incorporating known anomalous samples into training procedures of anomaly detection models. In this work, we propose a novel model-agnostic training procedure to address this task. We reformulate one-class classification as a binary classification problem with normal data being distinguished from pseudo-anomalous samples. The pseudo-anomalous samples are drawn from low-density regions of a normalizing flow model by feeding tails of the latent distribution into the model. Such an approach allows to easily include known anomalies into the training process of an arbitrary classifier. We demonstrate that our approach shows comparable performance on one-class problems, and, most importantly, achieves comparable or superior results on tasks with variable amounts of known anomalies.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.1912.09323</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Anomalies ; Bayesian analysis ; Computer Science - Learning ; Machine learning ; Normalizing (statistics) ; Probabilistic inference ; Statistical inference ; Statistics - Machine Learning</subject><ispartof>arXiv.org, 2021-11</ispartof><rights>2021. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27924</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.1912.09323$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.7717/peerj-cs.757$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Ryzhikov, Artem</creatorcontrib><creatorcontrib>Borisyak, Maxim</creatorcontrib><creatorcontrib>Ustyuzhanin, Andrey</creatorcontrib><creatorcontrib>Derkach, Denis</creatorcontrib><title>NFAD: Fixing anomaly detection using normalizing flows</title><title>arXiv.org</title><description>Anomaly detection is a challenging task that frequently arises in practically all areas of industry and science, from fraud detection and data quality monitoring to finding rare cases of diseases and searching for new physics. Most of the conventional approaches to anomaly detection, such as one-class SVM and Robust Auto-Encoder, are one-class classification methods, i.e. focus on separating normal data from the rest of the space. Such methods are based on the assumption of separability of normal and anomalous classes, and subsequently do not take into account any available samples of anomalies. Nonetheless, in practical settings, some anomalous samples are often available; however, usually in amounts far lower than required for a balanced classification task, and the separability assumption might not always hold. This leads to an important task -- incorporating known anomalous samples into training procedures of anomaly detection models. In this work, we propose a novel model-agnostic training procedure to address this task. We reformulate one-class classification as a binary classification problem with normal data being distinguished from pseudo-anomalous samples. The pseudo-anomalous samples are drawn from low-density regions of a normalizing flow model by feeding tails of the latent distribution into the model. Such an approach allows to easily include known anomalies into the training process of an arbitrary classifier. We demonstrate that our approach shows comparable performance on one-class problems, and, most importantly, achieves comparable or superior results on tasks with variable amounts of known anomalies.</description><subject>Anomalies</subject><subject>Bayesian analysis</subject><subject>Computer Science - Learning</subject><subject>Machine learning</subject><subject>Normalizing (statistics)</subject><subject>Probabilistic inference</subject><subject>Statistical inference</subject><subject>Statistics - Machine Learning</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotj1FLwzAUhYMgOOZ-gE8WfG5Nbpom8W1MO4WhL3svSZNKRtfMpNXNX2-7-XQPH4fD_RC6IzjLBWP4UYWj-86IJJBhSYFeoRlQSlKRA9ygRYw7jDEUHBijM1S8l8vnp6R0R9d9Jqrze9WeEmN7W_fOd8kQJ975MHL3O-Wm9T_xFl03qo128X_naFu-bFev6eZj_bZablLFQKRSq6KgFqgVDRFMEDCSG6V5zSWWpq455BZwoeVItFBGU8hzZcASrrUCOkf3l9mzVHUIbq_CqZrkqrPc2Hi4NA7Bfw029tXOD6Ebf6qAgpACUyzoH-slUSY</recordid><startdate>20211119</startdate><enddate>20211119</enddate><creator>Ryzhikov, Artem</creator><creator>Borisyak, Maxim</creator><creator>Ustyuzhanin, Andrey</creator><creator>Derkach, Denis</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20211119</creationdate><title>NFAD: Fixing anomaly detection using normalizing flows</title><author>Ryzhikov, Artem ; Borisyak, Maxim ; Ustyuzhanin, Andrey ; Derkach, Denis</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a528-9ba663e23e8f185812d97dab7c7909dcc724e206b97c7b8adb3244ad2e17bba23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Anomalies</topic><topic>Bayesian analysis</topic><topic>Computer Science - Learning</topic><topic>Machine learning</topic><topic>Normalizing (statistics)</topic><topic>Probabilistic inference</topic><topic>Statistical inference</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Ryzhikov, Artem</creatorcontrib><creatorcontrib>Borisyak, Maxim</creatorcontrib><creatorcontrib>Ustyuzhanin, Andrey</creatorcontrib><creatorcontrib>Derkach, Denis</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ryzhikov, Artem</au><au>Borisyak, Maxim</au><au>Ustyuzhanin, Andrey</au><au>Derkach, Denis</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>NFAD: Fixing anomaly detection using normalizing flows</atitle><jtitle>arXiv.org</jtitle><date>2021-11-19</date><risdate>2021</risdate><eissn>2331-8422</eissn><abstract>Anomaly detection is a challenging task that frequently arises in practically all areas of industry and science, from fraud detection and data quality monitoring to finding rare cases of diseases and searching for new physics. Most of the conventional approaches to anomaly detection, such as one-class SVM and Robust Auto-Encoder, are one-class classification methods, i.e. focus on separating normal data from the rest of the space. Such methods are based on the assumption of separability of normal and anomalous classes, and subsequently do not take into account any available samples of anomalies. Nonetheless, in practical settings, some anomalous samples are often available; however, usually in amounts far lower than required for a balanced classification task, and the separability assumption might not always hold. This leads to an important task -- incorporating known anomalous samples into training procedures of anomaly detection models. In this work, we propose a novel model-agnostic training procedure to address this task. We reformulate one-class classification as a binary classification problem with normal data being distinguished from pseudo-anomalous samples. The pseudo-anomalous samples are drawn from low-density regions of a normalizing flow model by feeding tails of the latent distribution into the model. Such an approach allows to easily include known anomalies into the training process of an arbitrary classifier. We demonstrate that our approach shows comparable performance on one-class problems, and, most importantly, achieves comparable or superior results on tasks with variable amounts of known anomalies.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.1912.09323</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2021-11
issn 2331-8422
language eng
recordid cdi_arxiv_primary_1912_09323
source arXiv.org; Free E- Journals
subjects Anomalies
Bayesian analysis
Computer Science - Learning
Machine learning
Normalizing (statistics)
Probabilistic inference
Statistical inference
Statistics - Machine Learning
title NFAD: Fixing anomaly detection using normalizing flows
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-11T09%3A40%3A35IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=NFAD:%20Fixing%20anomaly%20detection%20using%20normalizing%20flows&rft.jtitle=arXiv.org&rft.au=Ryzhikov,%20Artem&rft.date=2021-11-19&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.1912.09323&rft_dat=%3Cproquest_arxiv%3E2328980308%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2328980308&rft_id=info:pmid/&rfr_iscdi=true