Fast Shapley Value Estimation: A Unified Approach

Shapley values have emerged as a widely accepted and trustworthy tool, grounded in theoretical axioms, for addressing challenges posed by black-box models like deep neural networks. However, computing Shapley values encounters exponential complexity as the number of features increases. Various appro...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Borui, Tian, Baotong, Zheng, Wenzhao, Zhou, Jie, Lu, Jiwen
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 Zhang, Borui
Tian, Baotong
Zheng, Wenzhao
Zhou, Jie
Lu, Jiwen
description Shapley values have emerged as a widely accepted and trustworthy tool, grounded in theoretical axioms, for addressing challenges posed by black-box models like deep neural networks. However, computing Shapley values encounters exponential complexity as the number of features increases. Various approaches, including ApproSemivalue, KernelSHAP, and FastSHAP, have been explored to expedite the computation. In our analysis of existing approaches, we observe that stochastic estimators can be unified as a linear transformation of randomly summed values from feature subsets. Based on this, we investigate the possibility of designing simple amortized estimators and propose a straightforward and efficient one, SimSHAP, by eliminating redundant techniques. Extensive experiments conducted on tabular and image datasets validate the effectiveness of our SimSHAP, which significantly accelerates the computation of accurate Shapley values.
doi_str_mv 10.48550/arxiv.2311.01010
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2311_01010</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2311_01010</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2311_010103</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjY01DMwBEJOBkO3xOISheCMxIKc1EqFsMSc0lQF1-KSzNzEksz8PCsFR4XQvMy0zNQUBceCgqL8xOQMHgbWtMSc4lReKM3NIO_mGuLsoQs2PL6gCKi3qDIeZEk82BJjwioAIjEwKw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Fast Shapley Value Estimation: A Unified Approach</title><source>arXiv.org</source><creator>Zhang, Borui ; Tian, Baotong ; Zheng, Wenzhao ; Zhou, Jie ; Lu, Jiwen</creator><creatorcontrib>Zhang, Borui ; Tian, Baotong ; Zheng, Wenzhao ; Zhou, Jie ; Lu, Jiwen</creatorcontrib><description>Shapley values have emerged as a widely accepted and trustworthy tool, grounded in theoretical axioms, for addressing challenges posed by black-box models like deep neural networks. However, computing Shapley values encounters exponential complexity as the number of features increases. Various approaches, including ApproSemivalue, KernelSHAP, and FastSHAP, have been explored to expedite the computation. In our analysis of existing approaches, we observe that stochastic estimators can be unified as a linear transformation of randomly summed values from feature subsets. Based on this, we investigate the possibility of designing simple amortized estimators and propose a straightforward and efficient one, SimSHAP, by eliminating redundant techniques. Extensive experiments conducted on tabular and image datasets validate the effectiveness of our SimSHAP, which significantly accelerates the computation of accurate Shapley values.</description><identifier>DOI: 10.48550/arxiv.2311.01010</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Learning</subject><creationdate>2023-11</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/2311.01010$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2311.01010$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhang, Borui</creatorcontrib><creatorcontrib>Tian, Baotong</creatorcontrib><creatorcontrib>Zheng, Wenzhao</creatorcontrib><creatorcontrib>Zhou, Jie</creatorcontrib><creatorcontrib>Lu, Jiwen</creatorcontrib><title>Fast Shapley Value Estimation: A Unified Approach</title><description>Shapley values have emerged as a widely accepted and trustworthy tool, grounded in theoretical axioms, for addressing challenges posed by black-box models like deep neural networks. However, computing Shapley values encounters exponential complexity as the number of features increases. Various approaches, including ApproSemivalue, KernelSHAP, and FastSHAP, have been explored to expedite the computation. In our analysis of existing approaches, we observe that stochastic estimators can be unified as a linear transformation of randomly summed values from feature subsets. Based on this, we investigate the possibility of designing simple amortized estimators and propose a straightforward and efficient one, SimSHAP, by eliminating redundant techniques. Extensive experiments conducted on tabular and image datasets validate the effectiveness of our SimSHAP, which significantly accelerates the computation of accurate Shapley values.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjY01DMwBEJOBkO3xOISheCMxIKc1EqFsMSc0lQF1-KSzNzEksz8PCsFR4XQvMy0zNQUBceCgqL8xOQMHgbWtMSc4lReKM3NIO_mGuLsoQs2PL6gCKi3qDIeZEk82BJjwioAIjEwKw</recordid><startdate>20231102</startdate><enddate>20231102</enddate><creator>Zhang, Borui</creator><creator>Tian, Baotong</creator><creator>Zheng, Wenzhao</creator><creator>Zhou, Jie</creator><creator>Lu, Jiwen</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20231102</creationdate><title>Fast Shapley Value Estimation: A Unified Approach</title><author>Zhang, Borui ; Tian, Baotong ; Zheng, Wenzhao ; Zhou, Jie ; Lu, Jiwen</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2311_010103</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Borui</creatorcontrib><creatorcontrib>Tian, Baotong</creatorcontrib><creatorcontrib>Zheng, Wenzhao</creatorcontrib><creatorcontrib>Zhou, Jie</creatorcontrib><creatorcontrib>Lu, Jiwen</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhang, Borui</au><au>Tian, Baotong</au><au>Zheng, Wenzhao</au><au>Zhou, Jie</au><au>Lu, Jiwen</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Fast Shapley Value Estimation: A Unified Approach</atitle><date>2023-11-02</date><risdate>2023</risdate><abstract>Shapley values have emerged as a widely accepted and trustworthy tool, grounded in theoretical axioms, for addressing challenges posed by black-box models like deep neural networks. However, computing Shapley values encounters exponential complexity as the number of features increases. Various approaches, including ApproSemivalue, KernelSHAP, and FastSHAP, have been explored to expedite the computation. In our analysis of existing approaches, we observe that stochastic estimators can be unified as a linear transformation of randomly summed values from feature subsets. Based on this, we investigate the possibility of designing simple amortized estimators and propose a straightforward and efficient one, SimSHAP, by eliminating redundant techniques. Extensive experiments conducted on tabular and image datasets validate the effectiveness of our SimSHAP, which significantly accelerates the computation of accurate Shapley values.</abstract><doi>10.48550/arxiv.2311.01010</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2311.01010
ispartof
issn
language eng
recordid cdi_arxiv_primary_2311_01010
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
title Fast Shapley Value Estimation: A Unified Approach
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-08T23%3A19%3A44IST&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=Fast%20Shapley%20Value%20Estimation:%20A%20Unified%20Approach&rft.au=Zhang,%20Borui&rft.date=2023-11-02&rft_id=info:doi/10.48550/arxiv.2311.01010&rft_dat=%3Carxiv_GOX%3E2311_01010%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