{"type":"ScholarlyArticle","title":"Fairness without the Sensitive Attribute via Causal Variational Autoencoder","authors":[{"name":"Vincent Grari","slug":"vincent-grari","given":"Vincent","family":"Grari","affiliation":"Sorbonne Université , CNRS , ISIR , F-75005 Paris , France; AXA , Paris , France"},{"name":"Sylvain Lamprier","slug":"sylvain-lamprier","given":"Sylvain","family":"Lamprier","affiliation":"Sorbonne Université , CNRS , ISIR , F-75005 Paris , France"},{"name":"Marcin Detyniecki","slug":"marcin-detyniecki","given":"Marcin","family":"Detyniecki","affiliation":"AXA , Paris , France; Polish Academy of Science , IBS PAN , Warsaw , Poland"}],"abstract":"In recent years, most fairness strategies in machine learning have focused on mitigating unwanted biases by assuming that the sensitive information is available. However, in practice this is not always the case: due to privacy purposes and regulations such as RGPD in EU, many personal sensitive attributes are frequently not collected. Yet, only a few prior works address the issue of mitigating bias in such a difficult setting, in particular to meet classical fairness objectives such as Demographic Parity and Equalized Odds. By leveraging recent developments for approximate inference, we propose in this paper an approach to fill this gap. To infer a sensitive information proxy, we introduce a new variational auto-encoding-based framework named SRCVAE that relies on knowledge of the underlying causal graph. The bias mitigation is then done in an adversarial fairness approach. Our proposed method empirically achieves significant improvements over existing works in the field. We observe that the generated proxy’s latent space correctly recovers sensitive information and that our approach achieves a higher accuracy while obtaining the same level of fairness on two real datasets.","doi":"10.24963/ijcai.2022/98","url":"https://ijcai.org/proceedings/2022/98","pdf":"https://ijcai.org/proceedings/2022/0098.pdf","markdown":"https://ijcai.org/proceedings/2022/98.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/98","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"AI Ethics, Trust, Fairness","track":"Main Track","number":98,"pages":{"first":696,"last":702},"keywords":["AI Ethics, Trust, Fairness: Fairness & Diversity","Machine Learning: Bayesian Learning","Machine Learning: Autoencoders","AI Ethics, Trust, Fairness: Bias","Machine Learning: Representation learning"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986612","length_sec":90,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38984755","length_sec":354,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Adel et al., 2019] Tameem Adel, Isabel Valera, Zoubin Ghahra- mani, and Adrian Weller. One-network adversarial fairness. In AAAI'19, volume 33, pages 2412-2420, 2019.","title":"One-network adversarial fairness","authors":"Tameem Adel, Isabel Valera, Zoubin Ghahramani, Adrian Weller","year":"2019","venue":"AAAI'19","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Angwin et al., 2016] Julia Angwin, Jeff Larson, Surya Mattu, and Lauren Kirchner. Machine bias. ProPublica, May 23, 2016, 2016.","title":"Machine bias","authors":"Julia Angwin, Jeff Larson, Surya Mattu, Lauren Kirchner","year":"2016","venue":"ProPublica","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Bolukbasi et al., 2016] Tolga Bolukbasi, Kai-Wei Chang, James Y Zou, Venkatesh Saligrama, and Adam T Kalai. Man is to com- puter programmer as woman is to homemaker? debiasing word embeddings. In NeurIPS. Curran Associates, Inc., 2016.","title":"Man is to computer programmer as woman is to homemaker? debiasing word embeddings","authors":"Tolga Bolukbasi, Kai-Wei Chang, James Y Zou, Venkatesh Saligrama, Adam T Kalai","year":"2016","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Celis et al., 2019] L Elisa Celis, Lingxiao Huang, Vijay Keswani, and Nisheeth K Vishnoi. Classification with fairness constraints: A meta-algorithm with provable guarantees. In Proceedings of the Conference on Fairness, Accountability, and Transparency, pages 319-328, 2019.","title":"Classification with fairness constraints: A meta-algorithm with provable guarantees","authors":"Celis","year":"2019","venue":"Proceedings of the Conference on Fairness, Accountability, and Transparency","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Chen et al., 2016] Xi Chen, Diederik P Kingma, Tim Salimans, Yan Duan, Prafulla Dhariwal, John Schulman, Ilya Sutskever, and Pieter Abbeel. Variational lossy autoencoder. arXiv preprint arXiv:1611.02731, 2016.","title":"Variational lossy autoencoder","authors":"Xi Chen, P Diederik, Tim Kingma, Yan Salimans, Prafulla Duan, John Dhariwal, Ilya Schulman, Pieter Sutskever, Abbeel","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Chen et al., 2019] Jiahao Chen, Nathan Kallus, Xiaojie Mao, Ge- offry Svacha, and Madeleine Udell. Fairness under unaware- ness: Assessing disparity when protected class is unobserved. In Proceedings of the Conference on Fairness, Accountability, and Transparency, pages 339-348, 2019.","title":"Fairness under unawareness: Assessing disparity when protected class is unobserved","authors":"Jiahao Chen, Nathan Kallus, Xiaojie Mao, Geoffry Svacha, Madeleine Udell","year":"2019","venue":"Proceedings of the Conference on Fairness, Accountability, and Transparency","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Coston et al., 2019] Amanda Coston, Karthikeyan Natesan Rama- murthy, Dennis Wei, Kush R Varshney, Skyler Speakman, Zairah Mustahsan, and Supriyo Chakraborty. Fair transfer learning with missing protected attributes. In AAAI, 2019.","title":"Fair transfer learning with missing protected attributes","authors":"Amanda Coston, Karthikeyan Natesan Ramamurthy, Dennis Wei, R Kush, Skyler Varshney, Zairah Speakman, Supriyo Mustahsan, Chakraborty","year":"2019","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Dwork et al., 2012] Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, and Richard Zemel. Fairness through awareness. In Proceedings of the 3rd innovations in theoretical computer science conference, pages 214-226, 2012.","title":"Fairness through awareness","authors":"Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, Richard Zemel","year":"2012","venue":"Proceedings of the 3rd innovations in theoretical computer science conference","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Grari et al., 2020a] Vincent Grari, Sylvain Lamprier, and Marcin Detyniecki. Adversarial learning for counterfactual fairness. arXiv preprint arXiv:2008.13122, 2020.","title":"Adversarial learning for counterfactual fairness","authors":"Grari","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Grari et al., 2020b] Vincent Grari, Sylvain Lamprier, and Marcin Detyniecki. Fairness-aware neural rényi minimization for con- tinuous features. In Christian Bessiere, editor, Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelli- gence, IJCAI 2020, pages 2262-2268. ijcai.org, 2020.","title":"Fairness-aware neural rényi minimization for continuous features","authors":"Grari","year":"2020","venue":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI 2020","doi":null,"url":null,"internal_url":"/proceedings/2020/313"},{"position":11,"raw":"[Grari et al., 2021] Vincent Grari, Oualid El Hajouji, Sylvain Lam- prier, and Marcin Detyniecki. Learning unbiased representations via rényi minimization. ECML PKDD, 2021.","title":"Learning unbiased representations via rényi minimization","authors":"Vincent Grari, Oualid El Hajouji, Sylvain Lamprier, Marcin Detyniecki","year":"2021","venue":"ECML PKDD","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Gupta et al., 2018] Maya Gupta, Andrew Cotter, Mahdi Milani Fard, and Serena Wang. Proxy fairness. arXiv preprint arXiv:1806.11212, 2018.","title":"Proxy fairness","authors":"Maya Gupta, Andrew Cotter, Mahdi Milani Fard, Serena Wang","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Hardt et al., 2016] Moritz Hardt, Eric Price, and Nati Srebro. Equality of opportunity in supervised learning. Advances in neu- ral information processing systems, 29, 2016.","title":"Equality of opportunity in supervised learning","authors":"Moritz Hardt, Eric Price, Nati Srebro","year":"2016","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Hashimoto et al., 2018] Tatsunori Hashimoto, Megha Srivastava, Hongseok Namkoong, and Percy Liang. Fairness without demo- graphics in repeated loss minimization. In ICML, pages 1929- 1938. PMLR, 2018.","title":"Fairness without demographics in repeated loss minimization","authors":"Tatsunori Hashimoto, Megha Srivastava, Hongseok Namkoong, Percy Liang","year":"2018","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Kilbertus et al., 2018] Niki Kilbertus, Adrià Gascón, Matt Kusner, Michael Veale, Krishna Gummadi, and Adrian Weller. Blind jus- tice: Fairness with encrypted sensitive attributes. In ICML, 2018.","title":"Blind justice: Fairness with encrypted sensitive attributes","authors":"Niki Kilbertus, Adrià Gascón, Matt Kusner, Michael Veale, Krishna Gummadi, Adrian Weller","year":"2018","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kingma and Welling, 2013] Diederik P Kingma and Max Welling. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.","title":"Auto-encoding variational bayes","authors":"P Diederik, Max Kingma, Welling","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Lahoti et al., 2020] Preethi Lahoti, Alex Beutel, Jilin Chen, Kang Lee, Flavien Prost, Nithum Thain, Xuezhi Wang, and Ed H Chi. Fairness without demographics through adversarially reweighted learning. arXiv preprint arXiv:2006.13114, 2020.","title":"Fairness without demographics through adversarially reweighted learning","authors":"Preethi Lahoti, Alex Beutel, Jilin Chen, Kang Lee, Flavien Prost, Nithum Thain, Xuezhi Wang, Ed H Chi","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Lambrecht and E. Tucker, 2016] Anja Lambrecht and Catherine E. Tucker. Algorithmic Bias? An Empirical Study into Appar- ent Gender-Based Discrimination in the Display of STEM Career Ads. SSRN Electronic Journal, 2016.","title":"Algorithmic Bias? An Empirical Study into Apparent Gender-Based Discrimination in the Display of STEM Career Ads","authors":"E Lambrecht, Anja Tucker, Catherine E Lambrecht, Tucker","year":"2016","venue":"SSRN Electronic Journal","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Louizos et al., 2017] Christos Louizos, Uri Shalit, Joris M Mooij, David Sontag, Richard Zemel, and Max Welling. Causal effect inference with deep latent-variable models. In Advances in Neu- ral Information Processing Systems, pages 6446-6456, 2017.","title":"Causal effect inference with deep latent-variable models","authors":"Christos Louizos, Uri Shalit, M Joris, David Mooij, Richard Sontag, Max Zemel, Welling","year":"2017","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Madras et al., 2018] David Madras, Elliot Creager, Toniann Pitassi, and Richard Zemel. Learning adversarially fair and trans- ferable representations. In International Conference on Machine Learning, pages 3384-3393. PMLR, 2018.","title":"Learning adversarially fair and transferable representations","authors":"David Madras, Elliot Creager, Toniann Pitassi, Richard Zemel","year":"2018","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Mohri et al., 2019] Mehryar Mohri, Gary Sivek, and Ananda Theertha Suresh. Agnostic federated learning. In International Conference on Machine Learning, 2019.","title":"Agnostic federated learning","authors":"Mehryar Mohri, Gary Sivek, Ananda Theertha, Suresh","year":"2019","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Pedreshi et al., 2008] Dino Pedreshi, Salvatore Ruggieri, and Franco Turini. Discrimination-aware data mining. In KDD'08, page 560, 2008.","title":"Discrimination-aware data mining","authors":"Dino Pedreshi, Salvatore Ruggieri, Franco Turini","year":"2008","venue":"KDD'08","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Pfohl et al., 2019] Stephen Pfohl, Tony Duan, Daisy Yi Ding, and Nigam H Shah. Counterfactual reasoning for fair clinical risk prediction. arXiv preprint arXiv:1907.06260, 2019.","title":"Counterfactual reasoning for fair clinical risk prediction","authors":"Stephen Pfohl, Tony Duan, Daisy Yi Ding, H Nigam, Shah","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Rényi, 1959] Alfréd Rényi. On measures of dependence. Acta mathematica hungarica, 10(3-4):441-451, 1959.","title":"Alfréd Rényi. On measures of dependence","authors":"Rényi","year":"1959","venue":"Acta mathematica hungarica","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Schumann et al., 2019] Candice Schumann, Xuezhi Wang, Alex Beutel, Jilin Chen, Hai Qian, and Ed H Chi. Transfer of machine learning fairness across domains. arXiv preprint arXiv:1906.09688, 2019.","title":"Transfer of machine learning fairness across domains","authors":"Candice Schumann, Xuezhi Wang, Alex Beutel, Jilin Chen, Hai Qian, Ed H Chi","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Sønderby et al., 2016] Casper Kaae Sønderby, Tapani Raiko, Lars Maaløe, Søren Kaae Sønderby, and Ole Winther. Ladder varia- tional autoencoders. In NIPS'16, pages 3738-3746, 2016.","title":"Ladder variational autoencoders","authors":"Sønderby","year":"2016","venue":"NIPS'16","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Veale and Binns, 2017] Michael Veale and Reuben Binns. Fairer machine learning in the real world: Mitigating discrimina- tion without collecting sensitive data. Big Data & Society, 4(2):2053951717743530, 2017.","title":"Fairer machine learning in the real world: Mitigating discrimination without collecting sensitive data","authors":"Binns Veale, Michael Veale, Reuben Binns","year":"2017","venue":"Big Data & Society","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Wadsworth et al., 2018] Christina Wadsworth, Francesca Vera, and Chris Piech. Achieving fairness through adversarial learn- ing: an application to recidivism prediction. arXiv:1807.00199, 2018.","title":"Achieving fairness through adversarial learning: an application to recidivism prediction","authors":"Christina Wadsworth, Francesca Vera, Chris Piech","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Yan et al., 2020] Shen Yan, Hsien-te Kao, and Emilio Ferrara. Fair class balancing: enhancing model fairness without observing sen- sitive attributes. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management, 2020.","title":"Fair class balancing: enhancing model fairness without observing sensitive attributes","authors":"Hsien-Te Shen Yan, Emilio Kao, Ferrara","year":"2020","venue":"Proceedings of the 29th ACM International Conference on Information & Knowledge Management","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zafar et al., 2015] Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez Rodriguez, and Krishna P Gummadi. Fair- ness constraints: Mechanisms for fair classification. arXiv preprint arXiv:1507.05259, 2015.","title":"Fairness constraints: Mechanisms for fair classification","authors":"Zafar","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Zhang et al., 2018] Brian Hu Zhang, Blake Lemoine, and Mar- garet Mitchell. Mitigating unwanted biases with adversarial learning. In AAAI'18, pages 335-340, 2018.","title":"Mitigating unwanted biases with adversarial learning","authors":"Brian Hu Zhang, Blake Lemoine, Margaret Mitchell","year":"2018","venue":"AAAI'18","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zhao et al., 2017] Shengjia Zhao, Jiaming Song, and Stefano Er- mon. Infovae: Information maximizing variational autoencoders. arXiv preprint arXiv:1706.02262, 2017.","title":"Infovae: Information maximizing variational autoencoders","authors":"Zhao","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Zhao et al., 2021] Tianxiang Zhao, Enyan Dai, Kai Shu, and Suhang Wang. You can still achieve fairness without sensi- tive attributes: Exploring biases in non-sensitive features. arXiv preprint arXiv:2104.14537, 2021.","title":"You can still achieve fairness without sensitive attributes: Exploring biases in non-sensitive features","authors":"Tianxiang Zhao, Enyan Dai, Kai Shu, Suhang Wang","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}