Fairness in Decision-Making - The Causal Explanation Formula

AI plays an increasingly prominent role in society since decisions that were once made by humans are now delegated to automated systems. These systems are currently in charge of deciding bank loans, criminals’ incarceration, and the hiring of new employees, and it’s not difficult to envision that they will in the future underpin most of the decisions in society. Despite the high complexity entailed by this task, there is still not much understanding of basic properties of such systems. For instance, we currently cannot detect (neither explain nor correct) whether an AI system is operating fairly (i.e., is abiding by the decision-constraints agreed by society) or it is reinforcing biases and perpetuating a preceding prejudicial practice. Issues of discrimination have been discussed extensively in legal circles, but there exists still not much understanding of the formal conditions that a system must adhere to be deemed fair. In this paper, we use the language of structural causality (Pearl, 2000) to fill in this gap. We start by introducing three new fine-grained measures of transmission of change from stimulus to effect, which we called counterfactual direct (Ctf-DE), indirect (Ctf-IE), and spurious (Ctf-SE) effects. We then derive the causal explanation formula, which allows the AI designer to quantitatively evaluate fairness and explain the total observed disparity of decisions through different discriminatory mechanisms. We apply these results to various discrimination analysis tasks and run extensive simulations, including detection, evaluation, and optimization of decision-making under fairness constraints. We conclude studying the trade-off between different types of fairness criteria (outcome and procedural), and provide a quantitative approach to policy implementation and the design of fair decision-making systems.

[1]  Lu Zhang,et al.  A Causal Framework for Discovering and Removing Direct and Indirect Discrimination , 2016, IJCAI.

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

[3]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[4]  A. Lo,et al.  Consumer Credit Risk Models Via Machine-Learning Algorithms , 2010 .

[5]  J. Fox,et al.  Effect Analysis in Structural-Equation Models II , 1985 .

[6]  Kosuke Imai,et al.  Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies , 2011, American Political Science Review.

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

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

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

[10]  S. Wright,et al.  The Theory of Path Coefficients a Reply to Niles's Criticism. , 1923, Genetics.

[11]  Elias Bareinboim,et al.  Bandits with Unobserved Confounders: A Causal Approach , 2015, NIPS.

[12]  S. Wright.  The Method of Path Coefficients , 1934 .

[13]  Salvatore Ruggieri,et al.  A multidisciplinary survey on discrimination analysis , 2013, The Knowledge Engineering Review.

[14]  D P MacKinnon,et al.  The intermediate endpoint effect in logistic and probit regression , 2007, Clinical trials.

[15]  R. P. McDonald,et al.  Structural Equations with Latent Variables , 1989 .

[16]  Elias Bareinboim,et al.  Causal inference and the data-fusion problem , 2016, Proceedings of the National Academy of Sciences.

[17]  L. Keele,et al.  Identification, Inference and Sensitivity Analysis for Causal Mediation Effects , 2010, 1011.1079.

[18]  Joseph Y. Halpern.  Axiomatizing Causal Reasoning , 1998, UAI.

[19]  D. A. Kenny,et al.  The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. , 1986, Journal of personality and social psychology.

[20]  Andrew D. Selbst,et al.  Big Data's Disparate Impact , 2016 .

[21]  Chris Clifton,et al.  Combating discrimination using Bayesian networks , 2014, Artificial Intelligence and Law.

[22]  T. Brennan,et al.  Evaluating the Predictive Validity of the Compas Risk and Needs Assessment System , 2009 .

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

[24]  Matt J. Kusner,et al.  Counterfactual Fairness , 2017, NIPS.

[25]  J. Böhnke.  Explanation in causal inference: Methods for mediation and interaction. , 2016, Quarterly journal of experimental psychology.

[26]  D. Mackinnon.  Introduction to Statistical Mediation Analysis , 2008 .

[27]  Francesco Bonchi,et al.  Exposing the probabilistic causal structure of discrimination , 2015, International Journal of Data Science and Analytics.

[28]  Krishna P. Gummadi,et al.  Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment , 2016, WWW.

[29]  T. Lancaster.  Introduction to Structural Equation Models , 1976 .

[30]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[31]  Krishna P. Gummadi,et al.  Learning Fair Classifiers , 2015, 1507.05259.

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

[33]  Alexandra Chouldechova,et al.  Fair prediction with disparate impact: A study of bias in recidivism prediction instruments , 2016, Big Data.

[34]  J. Pearl.  The Mediation Formula: A Guide to the Assessment of Causal Pathways in Nonlinear Models , 2011 .

[35]  Ian Ayres.  Narrow Tailoring , 2019 .

[36]  Rebecca M. Blank,et al.  Race and gender in the labor market , 1999 .

[37]  Bernhard Schölkopf,et al.  Avoiding Discrimination through Causal Reasoning , 2017, NIPS.

[38]  J. Yinger,et al.  Measuring racial discrimination with fair housing audits: caught in the Act. , 1986 .

[39]  Toon Calders,et al.  Building Classifiers with Independency Constraints , 2009, 2009 IEEE International Conference on Data Mining Workshops.

[40]  Yair Zick,et al.  Algorithmic Transparency via Quantitative Input Influence: Theory and Experiments with Learning Systems , 2016, 2016 IEEE Symposium on Security and Privacy (SP).

[41]  Judea Pearl,et al.  Direct and Indirect Effects , 2001, UAI.