FairTest: Discovering Unwarranted Associations in Data-Driven Applications

In a world where traditional notions of privacy are increasingly challenged by the myriad companies that collect and analyze our data, it is important that decision-making entities are held accountable for unfair treatments arising from irresponsible data usage. Unfortunately, a lack of appropriate methodologies and tools means that even identifying unfair or discriminatory effects can be a challenge in practice. We introduce the unwarranted associations (UA) framework, a principled methodology for the discovery of unfair, discriminatory, or offensive user treatment in data-driven applications. The UA framework unifies and rationalizes a number of prior attempts at formalizing algorithmic fairness. It uniquely combines multiple investigative primitives and fairness metrics with broad applicability, granular exploration of unfair treatment in user subgroups, and incorporation of natural notions of utility that may account for observed disparities. We instantiate the UA framework in FairTest, the first comprehensive tool that helps developers check data-driven applications for unfair user treatment. It enables scalable and statistically rigorous investigation of associations between application outcomes (such as prices or premiums) and sensitive user attributes (such as race or gender). Furthermore, FairTest provides debugging capabilities that let programmers rule out potential confounders for observed unfair effects. We report on use of FairTest to investigate and in some cases address disparate impact, offensive labeling, and uneven rates of algorithmic error in four data-driven applications. As examples, our results reveal subtle biases against older populations in the distribution of error in a predictive health application and offensive racial labeling in an image tagger.

[1]  E. H. Simpson,et al.  The Interpretation of Interaction in Contingency Tables , 1951 .

[2]  F. James Rohlf,et al.  Biometry: The Principles and Practice of Statistics in Biological Research , 1969 .

[3]  P. Bickel,et al.  Sex Bias in Graduate Admissions: Data from Berkeley , 1975, Science.

[4]  S. Holm A Simple Sequentially Rejective Multiple Test Procedure , 1979 .

[5]  J. Rodgers,et al.  Thirteen ways to look at the correlation coefficient , 1988 .

[6]  David W. Hosmer,et al.  Applied Logistic Regression , 1991 .

[7]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[8]  Graham J. G. Upton,et al.  A Dictionary of Statistics , 2002 .

[9]  Liam Paninski,et al.  Estimation of Entropy and Mutual Information , 2003, Neural Computation.

[10]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[11]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[12]  M. D. Ernst Permutation Methods: A Basis for Exact Inference , 2004 .

[13]  Bernhard Schölkopf,et al.  Measuring Statistical Dependence with Hilbert-Schmidt Norms , 2005, ALT.

[14]  T. Carrilio Looking Inside the “Black Box” , 2006 .

[15]  A. Asuncion,et al.  UCI Machine Learning Repository, University of California, Irvine, School of Information and Computer Sciences , 2007 .

[16]  Teva J. Scheer Uniform Guidelines on Employee Selection Procedures , 2007 .

[17]  Maria L. Rizzo,et al.  Measuring and testing dependence by correlation of distances , 2007, 0803.4101.

[18]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

[19]  Franco Turini,et al.  Discrimination-aware data mining , 2008, KDD.

[20]  Jennifer L. Peresie Toward a Coherent Test for Disparate Impact Discrimination , 2009 .

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

[22]  Fei-Fei Li,et al.  ImageNet: A large-scale hierarchical image database , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[23]  Toon Calders,et al.  Three naive Bayes approaches for discrimination-free classification , 2010, Data Mining and Knowledge Discovery.

[24]  Franco Turini,et al.  Data mining for discrimination discovery , 2010, TKDD.

[25]  Toon Calders,et al.  Discrimination Aware Decision Tree Learning , 2010, 2010 IEEE International Conference on Data Mining.

[26]  Franco Turini,et al.  Integrating induction and deduction for finding evidence of discrimination , 2009, Artificial Intelligence and Law.

[27]  Toon Calders,et al.  Handling Conditional Discrimination , 2011, 2011 IEEE 11th International Conference on Data Mining.

[28]  Franco Turini,et al.  k-NN as an implementation of situation testing for discrimination discovery and prevention , 2011, KDD.

[29]  Barnabás Póczos,et al.  Nonparametric Divergence Estimation with Applications to Machine Learning on Distributions , 2011, UAI.

[30]  Jun Sakuma,et al.  Fairness-Aware Classifier with Prejudice Remover Regularizer , 2012, ECML/PKDD.

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

[32]  Epcd About - U.S. Census Bureau , 2012 .

[33]  Balachander Krishnamurthy,et al.  Measuring personalization of web search , 2013, WWW.

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

[35]  Latanya Sweeney,et al.  Discrimination in online ad delivery , 2013, CACM.

[36]  Sofya Raskhodnikova,et al.  Testing the Lipschitz Property over Product Distributions with Applications to Data Privacy , 2013, TCC.

[37]  Ramesh Govindan,et al.  AdReveal: improving transparency into online targeted advertising , 2013, HotNets.

[38]  Josep Domingo-Ferrer,et al.  A Methodology for Direct and Indirect Discrimination Prevention in Data Mining , 2013, IEEE Transactions on Knowledge and Data Engineering.

[39]  Roxana Geambasu,et al.  XRay: Enhancing the Web's Transparency with Differential Correlation , 2014, USENIX Security Symposium.

[40]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[41]  R. Fergus,et al.  OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks , 2013, ICLR.

[42]  Qiang Ma,et al.  Adscape: harvesting and analyzing online display ads , 2014, WWW.

[43]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[44]  Michael Carl Tschantz,et al.  Automated Experiments on Ad Privacy Settings: A Tale of Opacity, Choice, and Discrimination , 2014, ArXiv.

[45]  Wouter Joosen,et al.  Crying wolf? On the price discrimination of online airline tickets , 2014, PETS 2014.

[46]  Saikat Guha,et al.  Bootstrapping Privacy Compliance in Big Data Systems , 2014, 2014 IEEE Symposium on Security and Privacy.

[47]  David Lazer,et al.  Measuring Price Discrimination and Steering on E-commerce Web Sites , 2014, Internet Measurement Conference.

[48]  Patrick Seemann,et al.  Matrix Factorization Techniques for Recommender Systems , 2014 .

[49]  Trevor J. Hastie,et al.  Confidence intervals for random forests: the jackknife and the infinitesimal jackknife , 2013, J. Mach. Learn. Res..

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

[51]  F. Maxwell Harper,et al.  The MovieLens Datasets: History and Context , 2016, TIIS.

[52]  Michael Carl Tschantz,et al.  Automated Experiments on Ad Privacy Settings , 2014, Proc. Priv. Enhancing Technol..

[53]  Dan S. Wallach,et al.  An Empirical Study of Mobile Ad Targeting , 2015, ArXiv.

[54]  Toniann Pitassi,et al.  The reusable holdout: Preserving validity in adaptive data analysis , 2015, Science.

[55]  Roxana Geambasu,et al.  Sunlight: Fine-grained Targeting Detection at Scale with Statistical Confidence , 2015, CCS.

[56]  Kristin Branson,et al.  Understanding classifier errors by examining influential neighbors , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[57]  Ameet Talwalkar,et al.  MLlib: Machine Learning in Apache Spark , 2015, J. Mach. Learn. Res..

[58]  Kyle Kubler The Black Box Society: the secret algorithms that control money and information , 2016 .

[59]  J. Reidenberg,et al.  Accountable Algorithms , 2016 .

[60]  Yair Zick,et al.  Algorithmic Transparency via Quantitative Input Influence , 2017 .