Three Factors to Improve Out-of-Distribution Detection

In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data for fine-tuning has demonstrated encouraging performance. However, previous methods have suffered from a trade-off between classification accuracy (ACC) and OOD detection performance (AUROC, FPR, AUPR)...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-08
Main Authors: Choi, Hyunjun, Chung, JaeHo, Jeong, Hawook, Jin Young Choi
Format: Article
Language:English
Subjects:
Classification
Data analysis
Distillation
Outliers (statistics)
Performance measurement
Tradeoffs
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Choi, Hyunjun
Chung, JaeHo
Jeong, Hawook
Jin Young Choi
description In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data for fine-tuning has demonstrated encouraging performance. However, previous methods have suffered from a trade-off between classification accuracy (ACC) and OOD detection performance (AUROC, FPR, AUPR). To improve this trade-off, we make three contributions: (i) Incorporating a self-knowledge distillation loss can enhance the accuracy of the network; (ii) Sampling semi-hard outlier data for training can improve OOD detection performance with minimal impact on accuracy; (iii) The introduction of our novel supervised contrastive learning can simultaneously improve OOD detection performance and the accuracy of the network. By incorporating all three factors, our approach enhances both accuracy and OOD detection performance by addressing the trade-off between classification and OOD detection. Our method achieves improvements over previous approaches in both performance metrics.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2845344578</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2845344578</sourcerecordid><originalsourceid>FETCH-proquest_journals_28453445783</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwC8koSk1VcEtMLskvKlYoyVfwzC0oyi9LVfAvLdHNT9N1ySwuKcpMKi3JzM9TcEktSU0GsXgYWNMSc4pTeaE0N4Oym2uIs4cuUG9haWpxSXxWfmlRHlAq3sjCxNTYxMTU3MKYOFUAu8Y1VA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2845344578</pqid></control><display><type>article</type><title>Three Factors to Improve Out-of-Distribution Detection</title><source>Publicly Available Content Database</source><creator>Choi, Hyunjun ; Chung, JaeHo ; Jeong, Hawook ; Jin Young Choi</creator><creatorcontrib>Choi, Hyunjun ; Chung, JaeHo ; Jeong, Hawook ; Jin Young Choi</creatorcontrib><description>In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data for fine-tuning has demonstrated encouraging performance. However, previous methods have suffered from a trade-off between classification accuracy (ACC) and OOD detection performance (AUROC, FPR, AUPR). To improve this trade-off, we make three contributions: (i) Incorporating a self-knowledge distillation loss can enhance the accuracy of the network; (ii) Sampling semi-hard outlier data for training can improve OOD detection performance with minimal impact on accuracy; (iii) The introduction of our novel supervised contrastive learning can simultaneously improve OOD detection performance and the accuracy of the network. By incorporating all three factors, our approach enhances both accuracy and OOD detection performance by addressing the trade-off between classification and OOD detection. Our method achieves improvements over previous approaches in both performance metrics.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Classification ; Data analysis ; Distillation ; Outliers (statistics) ; Performance measurement ; Tradeoffs</subject><ispartof>arXiv.org, 2023-08</ispartof><rights>2023. 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><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2845344578?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>776,780,25731,36989,44566</link.rule.ids></links><search><creatorcontrib>Choi, Hyunjun</creatorcontrib><creatorcontrib>Chung, JaeHo</creatorcontrib><creatorcontrib>Jeong, Hawook</creatorcontrib><creatorcontrib>Jin Young Choi</creatorcontrib><title>Three Factors to Improve Out-of-Distribution Detection</title><title>arXiv.org</title><description>In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data for fine-tuning has demonstrated encouraging performance. However, previous methods have suffered from a trade-off between classification accuracy (ACC) and OOD detection performance (AUROC, FPR, AUPR). To improve this trade-off, we make three contributions: (i) Incorporating a self-knowledge distillation loss can enhance the accuracy of the network; (ii) Sampling semi-hard outlier data for training can improve OOD detection performance with minimal impact on accuracy; (iii) The introduction of our novel supervised contrastive learning can simultaneously improve OOD detection performance and the accuracy of the network. By incorporating all three factors, our approach enhances both accuracy and OOD detection performance by addressing the trade-off between classification and OOD detection. Our method achieves improvements over previous approaches in both performance metrics.</description><subject>Classification</subject><subject>Data analysis</subject><subject>Distillation</subject><subject>Outliers (statistics)</subject><subject>Performance measurement</subject><subject>Tradeoffs</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwC8koSk1VcEtMLskvKlYoyVfwzC0oyi9LVfAvLdHNT9N1ySwuKcpMKi3JzM9TcEktSU0GsXgYWNMSc4pTeaE0N4Oym2uIs4cuUG9haWpxSXxWfmlRHlAq3sjCxNTYxMTU3MKYOFUAu8Y1VA</recordid><startdate>20230802</startdate><enddate>20230802</enddate><creator>Choi, Hyunjun</creator><creator>Chung, JaeHo</creator><creator>Jeong, Hawook</creator><creator>Jin Young Choi</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>PTHSS</scope></search><sort><creationdate>20230802</creationdate><title>Three Factors to Improve Out-of-Distribution Detection</title><author>Choi, Hyunjun ; Chung, JaeHo ; Jeong, Hawook ; Jin Young Choi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28453445783</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Classification</topic><topic>Data analysis</topic><topic>Distillation</topic><topic>Outliers (statistics)</topic><topic>Performance measurement</topic><topic>Tradeoffs</topic><toplevel>online_resources</toplevel><creatorcontrib>Choi, Hyunjun</creatorcontrib><creatorcontrib>Chung, JaeHo</creatorcontrib><creatorcontrib>Jeong, Hawook</creatorcontrib><creatorcontrib>Jin Young Choi</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</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</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>ProQuest 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>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Choi, Hyunjun</au><au>Chung, JaeHo</au><au>Jeong, Hawook</au><au>Jin Young Choi</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Three Factors to Improve Out-of-Distribution Detection</atitle><jtitle>arXiv.org</jtitle><date>2023-08-02</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>In the problem of out-of-distribution (OOD) detection, the usage of auxiliary data as outlier data for fine-tuning has demonstrated encouraging performance. However, previous methods have suffered from a trade-off between classification accuracy (ACC) and OOD detection performance (AUROC, FPR, AUPR). To improve this trade-off, we make three contributions: (i) Incorporating a self-knowledge distillation loss can enhance the accuracy of the network; (ii) Sampling semi-hard outlier data for training can improve OOD detection performance with minimal impact on accuracy; (iii) The introduction of our novel supervised contrastive learning can simultaneously improve OOD detection performance and the accuracy of the network. By incorporating all three factors, our approach enhances both accuracy and OOD detection performance by addressing the trade-off between classification and OOD detection. Our method achieves improvements over previous approaches in both performance metrics.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-08
issn 2331-8422
language eng
recordid cdi_proquest_journals_2845344578
source Publicly Available Content Database
subjects Classification
Data analysis
Distillation
Outliers (statistics)
Performance measurement
Tradeoffs
title Three Factors to Improve Out-of-Distribution Detection
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-28T17%3A47%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Three%20Factors%20to%20Improve%20Out-of-Distribution%20Detection&rft.jtitle=arXiv.org&rft.au=Choi,%20Hyunjun&rft.date=2023-08-02&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2845344578%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_28453445783%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2845344578&rft_id=info:pmid/&rfr_iscdi=true