{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,14]],"date-time":"2026-02-14T04:58:08Z","timestamp":1771045088975,"version":"3.50.1"},"reference-count":44,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2021,5,26]],"date-time":"2021-05-26T00:00:00Z","timestamp":1621987200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["SIGKDD Explor. Newsl."],"published-print":{"date-parts":[[2021,5,26]]},"abstract":"<jats:p>Machine learning is being integrated into a growing number of critical systems with far-reaching impacts on society. Unexpected behaviour and unfair decision processes are coming under increasing scrutiny due to this widespread use and its theoretical considerations. Individuals, as well as organisations, notice, test, and criticize unfair results to hold model designers and deployers accountable. We offer a framework that assists these groups in mitigating unfair representations stemming from the training datasets. Our framework relies on two inter-operating adversaries to improve fairness. First, a model is trained with the goal of preventing the guessing of protected attributes' values while limiting utility losses. This first step optimizes the model's parameters for fairness. Second, the framework leverages evasion attacks from adversarial machine learning to generate new examples that will be misclassified. These new examples are then used to retrain and improve the model in the first step. These two steps are iteratively applied until a significant improvement in fairness is obtained. We evaluated our framework on well-studied datasets in the fairness literature - including COMPAS - where it can surpass other approaches concerning demographic parity, equality of opportunity and also the model's utility. We investigated the trade-offs between these targets in terms of model hyperparameters and also illustrated our findings on the subtle difficulties when mitigating unfairness and highlight how our framework can assist model designers.<\/jats:p>","DOI":"10.1145\/3468507.3468513","type":"journal-article","created":{"date-parts":[[2021,5,30]],"date-time":"2021-05-30T00:55:35Z","timestamp":1622336135000},"page":"32-41","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":22,"title":["Ethical Adversaries"],"prefix":"10.1145","volume":"23","author":[{"given":"Pieter","family":"Delobelle","sequence":"first","affiliation":[{"name":"KU Leuven, Leuven.AI, Leuven, Belgium"}]},{"given":"Paul","family":"Temple","sequence":"additional","affiliation":[{"name":"University of Namur, Namur, Belgium"}]},{"given":"Gilles","family":"Perrouin","sequence":"additional","affiliation":[{"name":"University of Namur, Namur, Belgium"}]},{"given":"Benoit","family":"Fr\u00e9nay","sequence":"additional","affiliation":[{"name":"University of Namur, Namur, Belgium"}]},{"given":"Patrick","family":"Heymans","sequence":"additional","affiliation":[{"name":"University of Namur, Namur, Belgium"}]},{"given":"Bettina","family":"Berendt","sequence":"additional","affiliation":[{"name":"KU Leuven, Leuven.AI, Leuven, Belgium"}]}],"member":"320","published-online":{"date-parts":[[2021,5,29]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"Adrian Weller. One-Network Adversarial Fairness. In AAAI Conference on Artificial Intelligence","author":"Adel Tameem","year":"2019","unstructured":"Tameem Adel , Isabel Valera , Zoubin Ghahramani , and Adrian Weller. One-Network Adversarial Fairness. In AAAI Conference on Artificial Intelligence , 2019 . Tameem Adel, Isabel Valera, Zoubin Ghahramani, and Adrian Weller. One-Network Adversarial Fairness. In AAAI Conference on Artificial Intelligence, 2019."},{"key":"e_1_2_1_2_1","volume-title":"ProPublica","author":"Angwin Julia","year":"2016","unstructured":"Julia Angwin and Jeff Larson . Machine bias : There's software used across the country to predict future criminals. and it's biased against blacks . ProPublica , 2016 . Julia Angwin and Jeff Larson. Machine bias: There's software used across the country to predict future criminals. and it's biased against blacks. ProPublica, 2016."},{"key":"e_1_2_1_3_1","volume-title":"Fairness and Machine Learning. fairmlbook.org","author":"Barocas Solon","year":"2019","unstructured":"Solon Barocas , Moritz Hardt , and Arvind Narayanan . Fairness and Machine Learning. fairmlbook.org , 2019 . Solon Barocas, Moritz Hardt, and Arvind Narayanan. Fairness and Machine Learning. fairmlbook.org, 2019."},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICB.2013.6613006"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2018.07.023"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.5555\/3042573.3042761"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-40994-3_25"},{"key":"e_1_2_1_8_1","volume-title":"May","author":"Blodgett Su Lin","year":"2020","unstructured":"Su Lin Blodgett , Solon Barocas , Hal Daum\u00e9 III, and Hanna Wallach. Language (Technology) is Power: A Critical Survey of \"Bias\" in NLP. arXiv:2005.14050 [cs] , May 2020 . Su Lin Blodgett, Solon Barocas, Hal Daum\u00e9 III, and Hanna Wallach. Language (Technology) is Power: A Critical Survey of \"Bias\" in NLP. arXiv:2005.14050 [cs], May 2020."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-010-0190-x"},{"key":"e_1_2_1_10_1","volume-title":"Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. arXiv:1703.00056","author":"Chouldechova Alexandra","year":"2017","unstructured":"Alexandra Chouldechova . Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. arXiv:1703.00056 , 2017 . Alexandra Chouldechova. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. arXiv:1703.00056, 2017."},{"key":"e_1_2_1_11_1","volume-title":"The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning. arXiv:1808.00023","author":"Corbett-Davies Sam","year":"2018","unstructured":"Sam Corbett-Davies and Sharad Goel . The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning. arXiv:1808.00023 , 2018 . Sam Corbett-Davies and Sharad Goel. The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning. arXiv:1808.00023, 2018."},{"key":"e_1_2_1_12_1","volume-title":"Ethical adversaries: Towards mitigating unfairness with adversarial machine learning","author":"Delobelle Pieter","year":"2020","unstructured":"Pieter Delobelle , Paul Temple , Gilles Perrouin , Beno\u00eet Fr\u00e9nay , Patrick Heymans , and Bettina Berendt . Ethical adversaries: Towards mitigating unfairness with adversarial machine learning , 2020 . Pieter Delobelle, Paul Temple, Gilles Perrouin, Beno\u00eet Fr\u00e9nay, Patrick Heymans, and Bettina Berendt. Ethical adversaries: Towards mitigating unfairness with adversarial machine learning, 2020."},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.5555\/3361338.3361361"},{"key":"e_1_2_1_14_1","volume-title":"COMPAS Risk Scales: Demonstrating Accuracy Equity and Predictive Parity","author":"Dieterich William","year":"2016","unstructured":"William Dieterich , Christina Mendoza , and Tim Brennan . COMPAS Risk Scales: Demonstrating Accuracy Equity and Predictive Parity . 2016 . William Dieterich, Christina Mendoza, and Tim Brennan. COMPAS Risk Scales: Demonstrating Accuracy Equity and Predictive Parity. 2016."},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1224"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/2090236.2090255"},{"key":"e_1_2_1_17_1","volume-title":"Censoring Representations with an Adversary. arXiv:1511.05897","author":"Edwards Harrison","year":"2015","unstructured":"Harrison Edwards and Amos Storkey . Censoring Representations with an Adversary. arXiv:1511.05897 , 2015 . Harrison Edwards and Amos Storkey. Censoring Representations with an Adversary. arXiv:1511.05897, 2015."},{"key":"e_1_2_1_18_1","first-page":"11","volume-title":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing","author":"Elazar Yanai","year":"1865","unstructured":"Yanai Elazar and Yoav Goldberg . Adversarial removal of demographic attributes from text data . In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing , pages 11 -- 21 , Brussels, Belgium, 10. ACL. doi: 10. 1865 3\/v1\/D18--1002. 10.18653\/v1 Yanai Elazar and Yoav Goldberg. Adversarial removal of demographic attributes from text data. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 11--21, Brussels, Belgium, 10. ACL. doi: 10.18653\/v1\/D18--1002."},{"key":"e_1_2_1_19_1","volume-title":"Runaway Feedback Loops in Predictive Policing. arXiv:1706.09847","author":"Ensign Danielle","year":"2017","unstructured":"Danielle Ensign , Sorelle A. Friedler , Scott Neville , Carlos Scheidegger , and Suresh Venkatasubramanian . Runaway Feedback Loops in Predictive Policing. arXiv:1706.09847 , 2017 . Danielle Ensign, Sorelle A. Friedler, Scott Neville, Carlos Scheidegger, and Suresh Venkatasubramanian. Runaway Feedback Loops in Predictive Policing. arXiv:1706.09847, 2017."},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/3411501.3419419"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/2783258.2783311"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.5555\/2946645.2946704"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.5555\/2969033.2969125"},{"key":"e_1_2_1_24_1","first-page":"51","volume-title":"AAAI","volume":"18","author":"Grgic-Hlaca Nina","year":"2018","unstructured":"Nina Grgic-Hlaca , Muhammad Bilal Zafar , Krishna P Gummadi , and Adrian Weller . Beyond distributive fairness in algorithmic decision making: Feature selection for procedurally fair learning . In AAAI , volume 18 , pages 51 -- 60 , 2018 . Nina Grgic-Hlaca, Muhammad Bilal Zafar, Krishna P Gummadi, and Adrian Weller. Beyond distributive fairness in algorithmic decision making: Feature selection for procedurally fair learning. In AAAI, volume 18, pages 51--60, 2018."},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2012.72"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.5555\/3157382.3157469"},{"key":"e_1_2_1_27_1","volume-title":"Charanpal Dhanjal, Laura C Stoddart, and Vlasios Vasileiou. Metrics and methods for a systematic comparison of fairness-aware machine learning algorithms. arXiv preprint arXiv:2010.03986","author":"Jones Gareth P","year":"2020","unstructured":"Gareth P Jones , James M Hickey , Pietro G Di Stefano , Charanpal Dhanjal, Laura C Stoddart, and Vlasios Vasileiou. Metrics and methods for a systematic comparison of fairness-aware machine learning algorithms. arXiv preprint arXiv:2010.03986 , 2020 . Gareth P Jones, James M Hickey, Pietro G Di Stefano, Charanpal Dhanjal, Laura C Stoddart, and Vlasios Vasileiou. Metrics and methods for a systematic comparison of fairness-aware machine learning algorithms. arXiv preprint arXiv:2010.03986, 2020."},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.5555\/3001460.3001502"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3351095.3372853"},{"key":"e_1_2_1_30_1","first-page":"33","article-title":"Fairness without demographics through adversarially reweighted learning","author":"Lahoti Preethi","year":"2020","unstructured":"Preethi Lahoti , Alex Beutel , Jilin Chen , Kang Lee , Flavien Prost , Nithum Thain , Xuezhi Wang , and Ed Chi . Fairness without demographics through adversarially reweighted learning . Advances in Neural Information Processing Systems , 33 , 2020 . Preethi Lahoti, Alex Beutel, Jilin Chen, Kang Lee, Flavien Prost, Nithum Thain, Xuezhi Wang, and Ed Chi. Fairness without demographics through adversarially reweighted learning. Advances in Neural Information Processing Systems, 33, 2020.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_1_31_1","volume-title":"Learning Adversarially Fair and Transferable Representations. arXiv, abs\/1802.06309","author":"Madras David","year":"2018","unstructured":"David Madras , Elliot Creager , Toniann Pitassi , and Richard S. Zemel . Learning Adversarially Fair and Transferable Representations. arXiv, abs\/1802.06309 , 2018 . David Madras, Elliot Creager, Toniann Pitassi, and Richard S. Zemel. Learning Adversarially Fair and Transferable Representations. arXiv, abs\/1802.06309, 2018."},{"key":"e_1_2_1_32_1","volume-title":"Questioning the assumptions behind fairness solutions. arXiv:1811.11293","author":"Overdorf Rebekah","year":"2018","unstructured":"Rebekah Overdorf , Bogdan Kulynych , Ero Balsa , Carmela Troncoso , and Seda G\u00fcrses . Questioning the assumptions behind fairness solutions. arXiv:1811.11293 , 2018 . Rebekah Overdorf, Bogdan Kulynych, Ero Balsa, Carmela Troncoso, and Seda G\u00fcrses. Questioning the assumptions behind fairness solutions. arXiv:1811.11293, 2018."},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/EuroSP.2016.36"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3052973.3053009"},{"key":"e_1_2_1_35_1","unstructured":"Emma\n      Pierson Sam\n      Corbett-Davies and \n      Sharad\n      Goel\n    .\n  Fast threshold tests for detecting discrimination\n  . In Amos Storkey and Fernando Perez-Cruz editors 21st International Conference on Artificial Intelligence and Statistics volume \n  84\n   of \n  Proceedings of Machine Learning Research pages \n  96\n  --\n  105 Playa Blanca Lanzarote Canary Islands 2018\n  . PMLR.  Emma Pierson Sam Corbett-Davies and Sharad Goel. Fast threshold tests for detecting discrimination. In Amos Storkey and Fernando Perez-Cruz editors 21st International Conference on Artificial Intelligence and Statistics volume 84 of Proceedings of Machine Learning Research pages 96--105 Playa Blanca Lanzarote Canary Islands 2018. PMLR."},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/DSAA.2018.00029"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3278721.3278742"},{"key":"e_1_2_1_38_1","volume-title":"Fairness GAN: Generating Datasets With Fairness Properties Using a Generative Adversarial Network. IBM Journal of Res. and Dev., page 12","author":"Sattigeri Prasanna","year":"2019","unstructured":"Prasanna Sattigeri , Samuel C Hoffman , Vijil Chenthamarakshan , and Kush R Varshney . Fairness GAN: Generating Datasets With Fairness Properties Using a Generative Adversarial Network. IBM Journal of Res. and Dev., page 12 , 2019 . Prasanna Sattigeri, Samuel C Hoffman, Vijil Chenthamarakshan, and Kush R Varshney. Fairness GAN: Generating Datasets With Fairness Properties Using a Generative Adversarial Network. IBM Journal of Res. and Dev., page 12, 2019."},{"key":"e_1_2_1_39_1","volume-title":"Poisoning attacks on algorithmic fairness. arXiv preprint arXiv:2004.07401","author":"Solans David","year":"2020","unstructured":"David Solans , Battista Biggio , and Carlos Castillo . Poisoning attacks on algorithmic fairness. arXiv preprint arXiv:2004.07401 , 2020 . David Solans, Battista Biggio, and Carlos Castillo. Poisoning attacks on algorithmic fairness. arXiv preprint arXiv:2004.07401, 2020."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3336294.3336309"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/3038912.3052660"},{"key":"e_1_2_1_42_1","volume-title":"Krishna P. Gummadi, and Adrian Weller. From Parity to Preference-Based Notions of Fairness in Classification. arXiv:1707.00010","author":"Zafar Muhammad Bilal","year":"2017","unstructured":"Muhammad Bilal Zafar , Isabel Valera , Manuel Gomez Rodriguez , Krishna P. Gummadi, and Adrian Weller. From Parity to Preference-Based Notions of Fairness in Classification. arXiv:1707.00010 , 2017 . Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez Rodriguez, Krishna P. Gummadi, and Adrian Weller. From Parity to Preference-Based Notions of Fairness in Classification. arXiv:1707.00010, 2017."},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.5555\/3042817.3042973"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/3278721.3278779"}],"container-title":["ACM SIGKDD Explorations Newsletter"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3468507.3468513","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3468507.3468513","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T20:17:18Z","timestamp":1750191438000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3468507.3468513"}},"subtitle":["Towards Mitigating Unfairness with Adversarial Machine Learning"],"short-title":[],"issued":{"date-parts":[[2021,5,26]]},"references-count":44,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,5,26]]}},"alternative-id":["10.1145\/3468507.3468513"],"URL":"https:\/\/doi.org\/10.1145\/3468507.3468513","relation":{},"ISSN":["1931-0145","1931-0153"],"issn-type":[{"value":"1931-0145","type":"print"},{"value":"1931-0153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,5,26]]},"assertion":[{"value":"2021-05-29","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}