Towards Fair Classifiers Without Sensitive Attributes: Exploring Biases in Related Features

Despite the rapid development and great success of machine learning models, extensive studies have exposed their disadvantage of inheriting latent discrimination and societal bias from the training data. This phenomenon hinders their adoption on high-stake applications. Thus, many efforts have been taken for developing fair machine learning models. Most of them require that sensitive attributes are available during training to learn fair models. However, in many real-world applications, it is usually infeasible to obtain the sensitive attributes due to privacy or legal issues, which challenges existing fair-ensuring strategies. Though the sensitive attribute of each data sample is unknown, we observe that there are usually some non-sensitive features in the training data that are highly correlated with sensitive attributes, which can be used to alleviate the bias. Therefore, in this paper, we study a novel problem of exploring features that are highly correlated with sensitive attributes for learning fair and accurate classifiers. We theoretically show that by minimizing the correlation between these related features and model prediction, we can learn a fair classifier. Based on this motivation, we propose a novel framework which simultaneously uses these related features for accurate prediction and enforces fairness. In addition, the model can dynamically adjust the regularization weight of each related feature to balance its contribution on model classification and fairness. Experimental results on real-world datasets demonstrate the effectiveness of the proposed model for learning fair models with high classification accuracy.

[1]  Turgay Celik,et al.  Statistical and machine learning models in credit scoring: A systematic literature survey , 2020, Appl. Soft Comput..

[2]  Toniann Pitassi,et al.  Fairness through awareness , 2011, ITCS '12.

[3]  Carlos Eduardo Scheidegger,et al.  Certifying and Removing Disparate Impact , 2014, KDD.

[4]  Hanghang Tong,et al.  InFoRM: Individual Fairness on Graph Mining , 2020, KDD.

[5]  Toniann Pitassi,et al.  Learning Fair Representations , 2013, ICML.

[6]  Blake Lemoine,et al.  Mitigating Unwanted Biases with Adversarial Learning , 2018, AIES.

[7]  Joseph Weiss,et al.  Ethical Implications of Bias in Machine Learning , 2018, HICSS.

[8]  Christopher T. Lowenkamp,et al.  False Positives, False Negatives, and False Analyses: A Rejoinder to "Machine Bias: There's Software Used across the Country to Predict Future Criminals. and It's Biased against Blacks" , 2016 .

[9]  Linda F. Wightman LSAC National Longitudinal Bar Passage Study. LSAC Research Report Series. , 1998 .

[10]  Kush R. Varshney,et al.  Fairness GAN , 2018, IBM Journal of Research and Development.

[11]  Richard G. Baraniuk,et al.  Fast Alternating Direction Optimization Methods , 2014, SIAM J. Imaging Sci..

[12]  Dragica Radosav,et al.  Deep Learning and Medical Diagnosis: A Review of Literature , 2018, Multimodal Technol. Interact..

[13]  Kristina Lerman,et al.  A Survey on Bias and Fairness in Machine Learning , 2019, ACM Comput. Surv..

[14]  W. Stewart,et al.  Gender differences in the experience of headache. , 1990, Social science & medicine.

[15]  Nathan Srebro,et al.  Equality of Opportunity in Supervised Learning , 2016, NIPS.

[16]  Jon M. Kleinberg,et al.  On Fairness and Calibration , 2017, NIPS.

[17]  Zhe Zhao,et al.  Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations , 2017, ArXiv.

[18]  Lu Zhang,et al.  Achieving Non-Discrimination in Data Release , 2016, KDD.

[19]  W. Gellert,et al.  The VNR concise encyclopedia of mathematics , 1977 .

[20]  Julia Rubin,et al.  Fairness Definitions Explained , 2018, 2018 IEEE/ACM International Workshop on Software Fairness (FairWare).

[21]  Toniann Pitassi,et al.  Flexibly Fair Representation Learning by Disentanglement , 2019, ICML.

[22]  Krishna P. Gummadi,et al.  Fairness Constraints: Mechanisms for Fair Classification , 2015, AISTATS.

[23]  Suhang Wang,et al.  Say No to the Discrimination: Learning Fair Graph Neural Networks with Limited Sensitive Attribute Information. , 2020 .

[24]  Lauren C. Porter,et al.  Toward a Demographic Understanding of Incarceration Disparities: Race, Ethnicity, and Age Structure , 2015, Journal of quantitative criminology.

[25]  Julie A. Shah,et al.  Fairness in Multi-Agent Sequential Decision-Making , 2014, NIPS.

[26]  Ed H. Chi,et al.  Fairness without Demographics through Adversarially Reweighted Learning , 2020, NeurIPS.

[27]  Percy Liang,et al.  Fairness Without Demographics in Repeated Loss Minimization , 2018, ICML.

[28]  Krishna P. Gummadi,et al.  Operationalizing Individual Fairness with Pairwise Fair Representations , 2019, Proc. VLDB Endow..

[29]  S. Tamang,et al.  Potential Biases in Machine Learning Algorithms Using Electronic Health Record Data , 2018, JAMA internal medicine.

[30]  Toon Calders,et al.  Classifying without discriminating , 2009, 2009 2nd International Conference on Computer, Control and Communication.

[31]  Amos J. Storkey,et al.  Censoring Representations with an Adversary , 2015, ICLR.

[32]  Robin D. Burke,et al.  Multisided Fairness for Recommendation , 2017, ArXiv.

[33]  Stefan Bauer,et al.  On the Fairness of Disentangled Representations , 2019, NeurIPS.

[34]  Ed H. Chi,et al.  Fairness in Recommendation Ranking through Pairwise Comparisons , 2019, KDD.

[35]  Toon Calders,et al.  Data preprocessing techniques for classification without discrimination , 2011, Knowledge and Information Systems.

[36]  Kush R. Varshney,et al.  Fair Transfer Learning with Missing Protected Attributes , 2019, AIES.

[37]  Max Welling,et al.  The Variational Fair Autoencoder , 2015, ICLR.