Dimensionless Anomaly Detection on Multivariate Streams with Variance Norm and Path Signature

In this paper, we propose a dimensionless anomaly detection method for multivariate streams. Our method is independent of the unit of measurement for the different stream channels, therefore dimensionless. We first propose the variance norm, a generalisation of Mahalanobis distance to handle infinit...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Shao, Zhen, Chan, Ryan Sze-Yin, Cochrane, Thomas, Foster, Peter, Lyons, Terry
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 Shao, Zhen
Chan, Ryan Sze-Yin
Cochrane, Thomas
Foster, Peter
Lyons, Terry
description In this paper, we propose a dimensionless anomaly detection method for multivariate streams. Our method is independent of the unit of measurement for the different stream channels, therefore dimensionless. We first propose the variance norm, a generalisation of Mahalanobis distance to handle infinite-dimensional feature space and singular empirical covariance matrix rigorously. We then combine the variance norm with the path signature, an infinite collection of iterated integrals that provide global features of streams, to propose SigMahaKNN, a method for anomaly detection on (multivariate) streams. We show that SigMahaKNN is invariant to stream reparametrisation, stream concatenation and has a graded discrimination power depending on the truncation level of the path signature. We implement SigMahaKNN as an open-source software, and perform extensive numerical experiments, showing significantly improved anomaly detection on streams compared to isolation forest and local outlier factors in applications ranging from language analysis, hand-writing analysis, ship movement paths analysis and univariate time-series analysis.
doi_str_mv 10.48550/arxiv.2006.03487
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2006_03487</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2006_03487</sourcerecordid><originalsourceid>FETCH-LOGICAL-a677-9afd427072cad74ed343f44fde8a115cb9c0a0600050ee1767ae3cbdc7cc6c163</originalsourceid><addsrcrecordid>eNotj8tqwzAQRbXpoqT9gK6qH7A7smTJWYakj0D6gITuiplI41Zgy0VW0uTv66SFCxfO4sBh7EZArqqyhDuMB7_PCwCdg1SVuWQfC99RGHwfWhoGPgt9h-2RLyiRTSPl4553bfJ7jB4T8XWKhN3Af3z64u8nGCzxlz52HIPjbzjitf8MmHaRrthFg-1A1_8_YZuH-838KVu9Pi7ns1WG2phsio1ThQFTWHRGkZNKNko1jioUorTbqQUEDQAlEAmjDZK0W2eNtdoKLSfs9k977qu_o-8wHutTZ33ulL9P_E9f</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Dimensionless Anomaly Detection on Multivariate Streams with Variance Norm and Path Signature</title><source>arXiv.org</source><creator>Shao, Zhen ; Chan, Ryan Sze-Yin ; Cochrane, Thomas ; Foster, Peter ; Lyons, Terry</creator><creatorcontrib>Shao, Zhen ; Chan, Ryan Sze-Yin ; Cochrane, Thomas ; Foster, Peter ; Lyons, Terry</creatorcontrib><description>In this paper, we propose a dimensionless anomaly detection method for multivariate streams. Our method is independent of the unit of measurement for the different stream channels, therefore dimensionless. We first propose the variance norm, a generalisation of Mahalanobis distance to handle infinite-dimensional feature space and singular empirical covariance matrix rigorously. We then combine the variance norm with the path signature, an infinite collection of iterated integrals that provide global features of streams, to propose SigMahaKNN, a method for anomaly detection on (multivariate) streams. We show that SigMahaKNN is invariant to stream reparametrisation, stream concatenation and has a graded discrimination power depending on the truncation level of the path signature. We implement SigMahaKNN as an open-source software, and perform extensive numerical experiments, showing significantly improved anomaly detection on streams compared to isolation forest and local outlier factors in applications ranging from language analysis, hand-writing analysis, ship movement paths analysis and univariate time-series analysis.</description><identifier>DOI: 10.48550/arxiv.2006.03487</identifier><language>eng</language><subject>Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2020-06</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/2006.03487$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2006.03487$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Shao, Zhen</creatorcontrib><creatorcontrib>Chan, Ryan Sze-Yin</creatorcontrib><creatorcontrib>Cochrane, Thomas</creatorcontrib><creatorcontrib>Foster, Peter</creatorcontrib><creatorcontrib>Lyons, Terry</creatorcontrib><title>Dimensionless Anomaly Detection on Multivariate Streams with Variance Norm and Path Signature</title><description>In this paper, we propose a dimensionless anomaly detection method for multivariate streams. Our method is independent of the unit of measurement for the different stream channels, therefore dimensionless. We first propose the variance norm, a generalisation of Mahalanobis distance to handle infinite-dimensional feature space and singular empirical covariance matrix rigorously. We then combine the variance norm with the path signature, an infinite collection of iterated integrals that provide global features of streams, to propose SigMahaKNN, a method for anomaly detection on (multivariate) streams. We show that SigMahaKNN is invariant to stream reparametrisation, stream concatenation and has a graded discrimination power depending on the truncation level of the path signature. We implement SigMahaKNN as an open-source software, and perform extensive numerical experiments, showing significantly improved anomaly detection on streams compared to isolation forest and local outlier factors in applications ranging from language analysis, hand-writing analysis, ship movement paths analysis and univariate time-series analysis.</description><subject>Computer Science - Learning</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj8tqwzAQRbXpoqT9gK6qH7A7smTJWYakj0D6gITuiplI41Zgy0VW0uTv66SFCxfO4sBh7EZArqqyhDuMB7_PCwCdg1SVuWQfC99RGHwfWhoGPgt9h-2RLyiRTSPl4553bfJ7jB4T8XWKhN3Af3z64u8nGCzxlz52HIPjbzjitf8MmHaRrthFg-1A1_8_YZuH-838KVu9Pi7ns1WG2phsio1ThQFTWHRGkZNKNko1jioUorTbqQUEDQAlEAmjDZK0W2eNtdoKLSfs9k977qu_o-8wHutTZ33ulL9P_E9f</recordid><startdate>20200605</startdate><enddate>20200605</enddate><creator>Shao, Zhen</creator><creator>Chan, Ryan Sze-Yin</creator><creator>Cochrane, Thomas</creator><creator>Foster, Peter</creator><creator>Lyons, Terry</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20200605</creationdate><title>Dimensionless Anomaly Detection on Multivariate Streams with Variance Norm and Path Signature</title><author>Shao, Zhen ; Chan, Ryan Sze-Yin ; Cochrane, Thomas ; Foster, Peter ; Lyons, Terry</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a677-9afd427072cad74ed343f44fde8a115cb9c0a0600050ee1767ae3cbdc7cc6c163</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Shao, Zhen</creatorcontrib><creatorcontrib>Chan, Ryan Sze-Yin</creatorcontrib><creatorcontrib>Cochrane, Thomas</creatorcontrib><creatorcontrib>Foster, Peter</creatorcontrib><creatorcontrib>Lyons, Terry</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>Shao, Zhen</au><au>Chan, Ryan Sze-Yin</au><au>Cochrane, Thomas</au><au>Foster, Peter</au><au>Lyons, Terry</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Dimensionless Anomaly Detection on Multivariate Streams with Variance Norm and Path Signature</atitle><date>2020-06-05</date><risdate>2020</risdate><abstract>In this paper, we propose a dimensionless anomaly detection method for multivariate streams. Our method is independent of the unit of measurement for the different stream channels, therefore dimensionless. We first propose the variance norm, a generalisation of Mahalanobis distance to handle infinite-dimensional feature space and singular empirical covariance matrix rigorously. We then combine the variance norm with the path signature, an infinite collection of iterated integrals that provide global features of streams, to propose SigMahaKNN, a method for anomaly detection on (multivariate) streams. We show that SigMahaKNN is invariant to stream reparametrisation, stream concatenation and has a graded discrimination power depending on the truncation level of the path signature. We implement SigMahaKNN as an open-source software, and perform extensive numerical experiments, showing significantly improved anomaly detection on streams compared to isolation forest and local outlier factors in applications ranging from language analysis, hand-writing analysis, ship movement paths analysis and univariate time-series analysis.</abstract><doi>10.48550/arxiv.2006.03487</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2006.03487
ispartof
issn
language eng
recordid cdi_arxiv_primary_2006_03487
source arXiv.org
subjects Computer Science - Learning
Statistics - Machine Learning
title Dimensionless Anomaly Detection on Multivariate Streams with Variance Norm and Path Signature
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-14T07%3A37%3A33IST&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=Dimensionless%20Anomaly%20Detection%20on%20Multivariate%20Streams%20with%20Variance%20Norm%20and%20Path%20Signature&rft.au=Shao,%20Zhen&rft.date=2020-06-05&rft_id=info:doi/10.48550/arxiv.2006.03487&rft_dat=%3Carxiv_GOX%3E2006_03487%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