Fair Tree Classifier using Strong Demographic Parity

When dealing with sensitive data in automated data-driven decision-making, an important concern is to learn predictors with high performance towards a class label, whilst minimising for the discrimination towards any sensitive attribute, like gender or race, induced from biased data. A few hybrid tr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Barata, António Pereira, Takes, Frank W, Herik, H. Jaap van den, Veenman, Cor J
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 Barata, António Pereira
Takes, Frank W
Herik, H. Jaap van den
Veenman, Cor J
description When dealing with sensitive data in automated data-driven decision-making, an important concern is to learn predictors with high performance towards a class label, whilst minimising for the discrimination towards any sensitive attribute, like gender or race, induced from biased data. A few hybrid tree optimisation criteria exist that combine classification performance and fairness. Although the threshold-free ROC-AUC is the standard for measuring traditional classification model performance, current fair tree classification methods mainly optimise for a fixed threshold on both the classification task as well as the fairness metric. In this paper, we propose a compound splitting criterion which combines threshold-free (i.e., strong) demographic parity with ROC-AUC termed SCAFF -- Splitting Criterion AUC for Fairness -- and easily extends to bagged and boosted tree frameworks. Our method simultaneously leverages multiple sensitive attributes of which the values may be multicategorical or intersectional, and is tunable with respect to the unavoidable performance-fairness trade-off. In our experiments, we demonstrate how SCAFF generates models with performance and fairness with respect to binary, multicategorical, and multiple sensitive attributes.
doi_str_mv 10.48550/arxiv.2110.09295
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2110_09295</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2110_09295</sourcerecordid><originalsourceid>FETCH-LOGICAL-a675-2bf4bc3b543d8ce132ffac2caa6a7285cccd2e7e86cd6961352ab2f81f397ed33</originalsourceid><addsrcrecordid>eNotzkFuwjAQhWFvukC0B2BVXyA0HseOs6xSKJWQqNTso8lkTC1BQROoyu2hwOqX3uLpU2pi8mkRnMtfUP7S7xTMZcgrqNxIFXNMohth1vUGhyHFxKKPQ_pZ66-D7C554-1uLbj_TqQ_UdLh9KgeIm4Gfrp3rJr5rKkX2XL1_lG_LjP0pcugi0VHtnOF7QOxsRAjEhCixxKCI6IeuOTgqfeVN9YBdhCDibYqubd2rJ5vt1d2u5e0RTm1__z2yrdnhYRAKg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Fair Tree Classifier using Strong Demographic Parity</title><source>arXiv.org</source><creator>Barata, António Pereira ; Takes, Frank W ; Herik, H. Jaap van den ; Veenman, Cor J</creator><creatorcontrib>Barata, António Pereira ; Takes, Frank W ; Herik, H. Jaap van den ; Veenman, Cor J</creatorcontrib><description>When dealing with sensitive data in automated data-driven decision-making, an important concern is to learn predictors with high performance towards a class label, whilst minimising for the discrimination towards any sensitive attribute, like gender or race, induced from biased data. A few hybrid tree optimisation criteria exist that combine classification performance and fairness. Although the threshold-free ROC-AUC is the standard for measuring traditional classification model performance, current fair tree classification methods mainly optimise for a fixed threshold on both the classification task as well as the fairness metric. In this paper, we propose a compound splitting criterion which combines threshold-free (i.e., strong) demographic parity with ROC-AUC termed SCAFF -- Splitting Criterion AUC for Fairness -- and easily extends to bagged and boosted tree frameworks. Our method simultaneously leverages multiple sensitive attributes of which the values may be multicategorical or intersectional, and is tunable with respect to the unavoidable performance-fairness trade-off. In our experiments, we demonstrate how SCAFF generates models with performance and fairness with respect to binary, multicategorical, and multiple sensitive attributes.</description><identifier>DOI: 10.48550/arxiv.2110.09295</identifier><language>eng</language><subject>Computer Science - Learning</subject><creationdate>2021-10</creationdate><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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,777,882</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2110.09295$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2110.09295$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Barata, António Pereira</creatorcontrib><creatorcontrib>Takes, Frank W</creatorcontrib><creatorcontrib>Herik, H. Jaap van den</creatorcontrib><creatorcontrib>Veenman, Cor J</creatorcontrib><title>Fair Tree Classifier using Strong Demographic Parity</title><description>When dealing with sensitive data in automated data-driven decision-making, an important concern is to learn predictors with high performance towards a class label, whilst minimising for the discrimination towards any sensitive attribute, like gender or race, induced from biased data. A few hybrid tree optimisation criteria exist that combine classification performance and fairness. Although the threshold-free ROC-AUC is the standard for measuring traditional classification model performance, current fair tree classification methods mainly optimise for a fixed threshold on both the classification task as well as the fairness metric. In this paper, we propose a compound splitting criterion which combines threshold-free (i.e., strong) demographic parity with ROC-AUC termed SCAFF -- Splitting Criterion AUC for Fairness -- and easily extends to bagged and boosted tree frameworks. Our method simultaneously leverages multiple sensitive attributes of which the values may be multicategorical or intersectional, and is tunable with respect to the unavoidable performance-fairness trade-off. In our experiments, we demonstrate how SCAFF generates models with performance and fairness with respect to binary, multicategorical, and multiple sensitive attributes.</description><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzkFuwjAQhWFvukC0B2BVXyA0HseOs6xSKJWQqNTso8lkTC1BQROoyu2hwOqX3uLpU2pi8mkRnMtfUP7S7xTMZcgrqNxIFXNMohth1vUGhyHFxKKPQ_pZ66-D7C554-1uLbj_TqQ_UdLh9KgeIm4Gfrp3rJr5rKkX2XL1_lG_LjP0pcugi0VHtnOF7QOxsRAjEhCixxKCI6IeuOTgqfeVN9YBdhCDibYqubd2rJ5vt1d2u5e0RTm1__z2yrdnhYRAKg</recordid><startdate>20211018</startdate><enddate>20211018</enddate><creator>Barata, António Pereira</creator><creator>Takes, Frank W</creator><creator>Herik, H. Jaap van den</creator><creator>Veenman, Cor J</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20211018</creationdate><title>Fair Tree Classifier using Strong Demographic Parity</title><author>Barata, António Pereira ; Takes, Frank W ; Herik, H. Jaap van den ; Veenman, Cor J</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a675-2bf4bc3b543d8ce132ffac2caa6a7285cccd2e7e86cd6961352ab2f81f397ed33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Barata, António Pereira</creatorcontrib><creatorcontrib>Takes, Frank W</creatorcontrib><creatorcontrib>Herik, H. Jaap van den</creatorcontrib><creatorcontrib>Veenman, Cor J</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Barata, António Pereira</au><au>Takes, Frank W</au><au>Herik, H. Jaap van den</au><au>Veenman, Cor J</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Fair Tree Classifier using Strong Demographic Parity</atitle><date>2021-10-18</date><risdate>2021</risdate><abstract>When dealing with sensitive data in automated data-driven decision-making, an important concern is to learn predictors with high performance towards a class label, whilst minimising for the discrimination towards any sensitive attribute, like gender or race, induced from biased data. A few hybrid tree optimisation criteria exist that combine classification performance and fairness. Although the threshold-free ROC-AUC is the standard for measuring traditional classification model performance, current fair tree classification methods mainly optimise for a fixed threshold on both the classification task as well as the fairness metric. In this paper, we propose a compound splitting criterion which combines threshold-free (i.e., strong) demographic parity with ROC-AUC termed SCAFF -- Splitting Criterion AUC for Fairness -- and easily extends to bagged and boosted tree frameworks. Our method simultaneously leverages multiple sensitive attributes of which the values may be multicategorical or intersectional, and is tunable with respect to the unavoidable performance-fairness trade-off. In our experiments, we demonstrate how SCAFF generates models with performance and fairness with respect to binary, multicategorical, and multiple sensitive attributes.</abstract><doi>10.48550/arxiv.2110.09295</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2110.09295
ispartof
issn
language eng
recordid cdi_arxiv_primary_2110_09295
source arXiv.org
subjects Computer Science - Learning
title Fair Tree Classifier using Strong Demographic Parity
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-17T14%3A51%3A19IST&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=Fair%20Tree%20Classifier%20using%20Strong%20Demographic%20Parity&rft.au=Barata,%20Ant%C3%B3nio%20Pereira&rft.date=2021-10-18&rft_id=info:doi/10.48550/arxiv.2110.09295&rft_dat=%3Carxiv_GOX%3E2110_09295%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