More Web Proxy on the site http://driver.im/

research-article

Public Access

A Geometric Solution to Fair Representations

Authors:

Keith Burghardt,

Kristina LermanAuthors Info & Claims

AIES '20: Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society

Pages 279 - 285

https://doi.org/10.1145/3375627.3375864

Published: 07 February 2020 Publication History

Abstract

To reduce human error and prejudice, many high-stakes decisions have been turned over to machine algorithms. However, recent research suggests that this does not remove discrimination, and can perpetuate harmful stereotypes. While algorithms have been developed to improve fairness, they typically face at least one of three shortcomings: they are not interpretable, their prediction quality deteriorates quickly compared to unbiased equivalents, and %the methodology cannot easily extend other algorithms they are not easily transferable across models% (e.g., methods to reduce bias in random forests cannot be extended to neural networks) . To address these shortcomings, we propose a geometric method that removes correlations between data and any number of protected variables. Further, we can control the strength of debiasing through an adjustable parameter to address the trade-off between prediction quality and fairness. The resulting features are interpretable and can be used with many popular models, such as linear regression, random forest, and multilayer perceptrons. The resulting predictions are found to be more accurate and fair compared to several state-of-the-art fair AI algorithms across a variety of benchmark datasets. Our work shows that debiasing data is a simple and effective solution toward improving fairness.

References

[1]

Julia Angwin, Jeff Larson, Surya Mattu, and Lauren Kirchner. 2016. Machine Bias. https://www.propublica.org/article/machine-bias-risk-assessments-incriminal- sentencing

[2]

Richard Berk, Hoda Heidari, Shahin Jabbari, Matthew Joseph, Michael Kearns, Jamie Morgenstern, Seth Neel, and Aaron Roth. 2017. A Convex Framework for Fair Regression. (2017), 1--15. arXiv:1706.02409 http://arxiv.org/abs/1706.02409

[3]

Leo Breiman. 2001. Random forests. Machine learning 45, 1 (2001), 5--32.

Digital Library

[4]

Chih-Chung Chang and Chih-Jen Lin. 2011. LIBSVM: A library for support vector machines. ACM TIST 2, 3 (2011), 27.

[5]

Irene Chen, Fredrik D. Johansson, and David Sontag. 2018. Why Is My Classifier Discriminatory? (2018). https://doi.org/arXiv:1805.12002v2 arXiv:1805.12002

[6]

Alexandra Chouldechova. 2017. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big data 5, 2 (2017), 153--163.

[7]

Alexandra Chouldechova and Aaron Roth. 2018. The frontiers of fairness in machine learning. arXiv preprint arXiv:1810.08810 (2018).

[8]

D. Ciregan, U. Meier, and J. Schmidhuber. 2012. Multi-column deep neural networks for image classification. In CVPR. 3642--3649. https://doi.org/10.1109/ CVPR.2012.6248110

[9]

Shai Danziger, Jonathan Levav, and Liora Avnaim-Pesso. 2011. Extraneous factors in judicial decisions. PNAS 108, 17 (2011), 6889--6892. https://doi.org/10.1073/ pnas.1018033108

[10]

Julia Dressel and Hany Farid. 2018. The accuracy, fairness, and limits of predicting recidivism. Science advances 4, 1 (2018), eaao5580.

[11]

Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, and Richard Zemel. 2012. Fairness through awareness. In ITCS. ACM, 214--226.

[12]

Yoav Freund and Robert E Schapire. 1997. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences 55, 1 (1997), 119--139.

Digital Library

[13]

Ben Hutchinson and Margaret Mitchell. 2019. 50 Years of Test (Un) fairness: Lessons for Machine Learning. In FAT*. ACM, 49--58.

[14]

Ayush Jaiswal, Yue Wu, Wael AbdAlmageed, and Premkumar Natarajan. 2018. Unsupervised Adversarial Invariance. In NIPS. Curran Associates, Inc., 5092--5102. arXiv:1809.10083 http://arxiv.org/abs/1809.10083

[15]

James E. Johndrow and Kristian Lum. 2017. An algorithm for removing sensitive information: application to race-independent recidivism prediction. (2017), 1--25. arXiv:1703.04957 http://arxiv.org/abs/1703.04957

[16]

F Kamiran, T Calders, and M Pechenizkiy. 2010. Discrimination Aware Decision Tree Learning. In ICDM. 869--874. https://doi.org/10.1109/ICDM.2010.50

[17]

Jon Kleinberg, Sendhil Mullainathan, and Manish Raghavan. 2016. Inherent Trade- Offs in the Fair Determination of Risk Scores. (sep 2016), 1--23. arXiv:1609.05807 http://arxiv.org/abs/1609.05807

[18]

Arie W. Kruglanski and Icek Ajzen. 1983. Bias and error in human judgment. European Journal of Social Psychology 13, 1 (1983), 1--44. https://doi.org/10.1002/ejsp.2420130102 arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1002/ejsp.2420130102

[19]

Bo Liu, Ying Wei, and Yu Zhang amd Qiang Yang. 2017. Deep Neural Networks for High Dimension, Low Sample Size Data. In IJCAI. 2287--2293.

[20]

Francesco Locatello, Gabriele Abbati, Tom Rainforth, Stefan Bauer, Bernhard Schölkopf, and Olivier Bachem. 2019. On the Fairness of Disentangled Representations. arXiv preprint arXiv:1905.13662 (2019).

[21]

Christos Louizos, Kevin Swersky, Yujia Li, Max Welling, and Richard Zemel. 2015. The Variational Fair Autoencoder. (2015), 1--11. arXiv:1511.00830 http: //arxiv.org/abs/1511.00830

[22]

Anandi Mani, Sendhil Mullainathan, Eldar Shafir, and Jiaying Zhao. 2013. Poverty impedes cognitive function. science 341, 6149 (2013), 976--980.

[23]

Daniel Moyer, Shuyang Gao, Rob Brekelmans, Greg Ver Steeg, and Aram Galstyan. 2018. Invariant Representations without Adversarial Training. Nips (2018). arXiv:1805.09458 http://arxiv.org/abs/1805.09458

[24]

Matt Olfat and Anil Aswani. 2018. Convex Formulations for Fair Principal Component Analysis. (2018). arXiv:1802.03765 http://arxiv.org/abs/1802.03765

[25]

Matthew Olson, Abraham J. Wyner, and Richard Berk. 2018. Modern Neural Networks Generalize on Small Data Sets. In NIPS. 3623--3632.

[26]

Cathy O'Neil. 2016. Weapons of math destruction: How big data increases inequality and threatens democracy. Broadway Books.

[27]

Emma Pierson, Camelia Simoiu, Jan Overgoor, Sam Corbett-Davies, Vignesh Ramachandran, Cheryl Phillips, and Sharad Goel. 2017. A large-scale analysis of racial disparities in police stops across the United States. preprint arXiv:1706.05678 (2017).

[28]

Frank Rosenblatt. 1961. Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms. Spartan Books, Washington DC.

[29]

Samira Samadi, Uthaipon Tantipongpipat, Jamie Morgenstern, Mohit Singh, and Santosh Vempala. 2018. The Price of Fair PCA: One Extra Dimension. Nips (2018). https://doi.org/10.1152/ajprenal.00633.2017 arXiv:1811.00103

[30]

Anuj K Shah, Sendhil Mullainathan, and Eldar Shafir. 2012. Some consequences of having too little. Science 338, 6107 (2012), 682--685.

[31]

Sahil Verma and Julia Rubin. 2018. Fairness definitions explained. In 2018 IEEE/ACM International Workshop on Software Fairness (FairWare). IEEE, 1--7.

Digital Library

[32]

Qizhe Xie, Zihang Dai, Yulun Du, Eduard Hovy, and Graham Neubig. 2017. Controllable invariance through adversarial feature learning. NIPS 2017-December, Mmd (2017), 586--597. arXiv:arXiv:1705.11122v3

[33]

Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez Rodriguez, and Krishna P. Gummadi. 2015. Fairness Constraints: Mechanisms for Fair Classification. 54 (2015). https://doi.org/10.1109/TRO.2009.2019886 arXiv:1507.05259

[34]

Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez Rodriguez, and Krishna P. Gummadi. 2016. Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment. (2016). https://doi.org/ 10.1145/3038912.3052660 arXiv:1610.08452

[35]

Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez Rodriguez, Krishna P. Gummadi, and Adrian Weller. 2017. From Parity to Preference-based Notions of Fairness in Classification. In NIPS. arXiv:1707.00010 http://arxiv.org/abs/1707. 00010

[36]

Rich Zemel, Yu Wu, Kevin Swersky, Toni Pitassi, and Cynthia Dwork. 2013. Learning Fair Representations. In ICML. 325--333. http://proceedings.mlr.press/ v28/zemel13.html

Cited By

Shahbazi NSintos SAsudeh A(2024)FairHash: A Fair and Memory/Time-efficient HashmapProceedings of the ACM on Management of Data10.1145/36549392:3(1-29)Online publication date: 30-May-2024
https://dl.acm.org/doi/10.1145/3654939
Saarela MHong JPark J(2024)On the relation of causality- versus correlation-based feature selection on model fairnessProceedings of the 39th ACM/SIGAPP Symposium on Applied Computing10.1145/3605098.3636018(56-64)Online publication date: 8-Apr-2024
https://dl.acm.org/doi/10.1145/3605098.3636018
Huang YAlvernaz SKim SMaki PDai YPeñalver Bernabé B(2024)Predicting prenatal depression and assessing model bias using machine learning modelsBiological Psychiatry Global Open Science10.1016/j.bpsgos.2024.100376(100376)Online publication date: Aug-2024
https://doi.org/10.1016/j.bpsgos.2024.100376
Show More Cited By

Index Terms

A Geometric Solution to Fair Representations

Recommendations

Learning fair representations for accuracy parity
Abstract
With the application of artificial intelligence in various fields of society, algorithm fairness has become a concern in social decision-making. Researchers have proposed various fair algorithms to achieve statistical parity and equal ...
Highlights
- Analyzing the representations’ differences between sensitive groups.
- Proposing ...
T²-fair: a two-tiered time and throughput fair scheduler for multi-rate WLANs
MSWiM '06: Proceedings of the 9th ACM international symposium on Modeling analysis and simulation of wireless and mobile systems

Low throughput due to unfairness is a key problem in multi-rate wireless local area networks. To promote fairness and hence throughput, T²-Fair groups flows according to their average data rate, provides each group fair time allocations and ensures ...
Can CSMA/CA networks be made fair?
MobiCom '08: Proceedings of the 14th ACM international conference on Mobile computing and networking

We demonstrate that CSMA/CA networks, including IEEE 802.11 networks, exhibit severe fairness problem in many scenarios, where some hosts obtain most of the channel's bandwidth while others starve. Most existing solutions require nodes to overhear ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image ACM Conferences

AIES '20: Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society

February 2020

439 pages

ISBN:9781450371100

DOI:10.1145/3375627

General Chairs:
Annette Markham
Aarhus University | Loyola University
,
Julia Powles
University of Western Australia
,
Toby Walsh
TU Berlin | University of New South Wales | Data61
,
Anne L. Washington
New York University

Copyright © 2020 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

SIGAI: ACM Special Interest Group on Artificial Intelligence

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 07 February 2020

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Funding Sources

Conference

AIES '20

Sponsor:

SIGAI

AIES '20: AAAI/ACM Conference on AI, Ethics, and Society

February 7 - 9, 2020

NY, New York, USA

Acceptance Rates

Overall Acceptance Rate 61 of 162 submissions, 38%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

11
Total Citations
View Citations
1,312
Total Downloads

Downloads (Last 12 months)276
Downloads (Last 6 weeks)21

Reflects downloads up to 01 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Shahbazi NSintos SAsudeh A(2024)FairHash: A Fair and Memory/Time-efficient HashmapProceedings of the ACM on Management of Data10.1145/36549392:3(1-29)Online publication date: 30-May-2024
https://dl.acm.org/doi/10.1145/3654939
Saarela MHong JPark J(2024)On the relation of causality- versus correlation-based feature selection on model fairnessProceedings of the 39th ACM/SIGAPP Symposium on Applied Computing10.1145/3605098.3636018(56-64)Online publication date: 8-Apr-2024
https://dl.acm.org/doi/10.1145/3605098.3636018
Huang YAlvernaz SKim SMaki PDai YPeñalver Bernabé B(2024)Predicting prenatal depression and assessing model bias using machine learning modelsBiological Psychiatry Global Open Science10.1016/j.bpsgos.2024.100376(100376)Online publication date: Aug-2024
https://doi.org/10.1016/j.bpsgos.2024.100376
Vainio-Pekka HAgbese MJantunen MVakkuri VMikkonen TRousi RAbrahamsson P(2023)The Role of Explainable AI in the Research Field of AI EthicsACM Transactions on Interactive Intelligent Systems10.1145/359997413:4(1-39)Online publication date: 1-Jun-2023
https://dl.acm.org/doi/10.1145/3599974
Matias AZipitria I(2023)Promoting Ethical Uses in Artificial Intelligence Applied to EducationAugmented Intelligence and Intelligent Tutoring Systems10.1007/978-3-031-32883-1_53(604-615)Online publication date: 22-May-2023
https://doi.org/10.1007/978-3-031-32883-1_53
Fabris AMessina SSilvello GSusto G(2022)Tackling Documentation Debt: A Survey on Algorithmic Fairness DatasetsProceedings of the 2nd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization10.1145/3551624.3555286(1-13)Online publication date: 6-Oct-2022
https://dl.acm.org/doi/10.1145/3551624.3555286
Sheng YYang JWu YMao KShi YHu JJiang WYang LOshana R(2022)The larger the fairer?Proceedings of the 59th ACM/IEEE Design Automation Conference10.1145/3489517.3530427(163-168)Online publication date: 10-Jul-2022
https://dl.acm.org/doi/10.1145/3489517.3530427
Fabris AMessina SSilvello GSusto G(2022)Algorithmic fairness datasets: the story so farData Mining and Knowledge Discovery10.1007/s10618-022-00854-z36:6(2074-2152)Online publication date: 17-Sep-2022
https://doi.org/10.1007/s10618-022-00854-z
Mehrabi NMorstatter FSaxena NLerman KGalstyan A(2021)A Survey on Bias and Fairness in Machine LearningACM Computing Surveys10.1145/345760754:6(1-35)Online publication date: 13-Jul-2021
https://dl.acm.org/doi/10.1145/3457607
Corrales-Barquero RMarin-Raventos GBarrantes E(2021)A Review of Gender Bias Mitigation in Credit Scoring Models2021 Ethics and Explainability for Responsible Data Science (EE-RDS)10.1109/EE-RDS53766.2021.9708589(1-10)Online publication date: 27-Oct-2021
https://doi.org/10.1109/EE-RDS53766.2021.9708589
Show More Cited By

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Media

Figures

Other

Tables

View Table of Contents