Causal inference for social discrimination reasoning

  title={Causal inference for social discrimination reasoning},
  author={Bilal Qureshi and Faisal Kamiran and Asim Karim and Salvatore Ruggieri and Dino Pedreschi},
  journal={Journal of Intelligent Information Systems},
The discovery of discriminatory bias in human or automated decision making is a task of increasing importance and difficulty, exacerbated by the pervasive use of machine learning and data mining. Currently, discrimination discovery largely relies upon correlation analysis of decisions records, disregarding the impact of confounding biases. We present a method for causal discrimination discovery based on propensity score analysis , a statistical tool for filtering out the effect of confounding… 

Through the Data Management Lens: Experimental Analysis and Evaluation of Fair Classification

A broad analysis of 13 fair classification approaches and additional variants is contributed, over their correctness, fairness, efficiency, scalability, robustness to data errors, sensitivity to underlying ML model, data efficiency, and stability using a variety of metrics and real-world datasets.

A Bayesian Causal Inference Approach for Assessing Fairness in Clinical Decision-Making

A Bayesian causal inference approach for assessing a causal fairness notion called principal fairness in clinical settings is proposed and demonstrated using both simulated data and electronic health records (EHR) data.

The Big Three: A Methodology to Increase Data Science ROI by Answering the Questions Companies Care About

This paper proposes a methodology for categorizing and answering 'The Big Three' questions (what is going on, what is causing it, and what actions can I take that will optimize what I care about) using data science.

Convex Fairness Constrained Model Using Causal Effect Estimators

Novel models are devised, called FairCEEs, which remove discrimination while keeping explanatory bias, based on estimators of causal effect utilizing propensity score analysis and it is proved that Fair CEEs with the squared loss theoretically outperform a naive MD constraint model.



6. A Diagnostic Routine for the Detection of Consequential Heterogeneity of Causal Effects

Least squares regression estimates of causal effects are conditional-variance-weighted estimates of individual-level causal effects. In this paper, we extract from the literature on counterfactual

The central role of the propensity score in observational studies for causal effects

Abstract : The results of observational studies are often disputed because of nonrandom treatment assignment. For example, patients at greater risk may be overrepresented in some treatment group.

Fairness in Criminal Justice Risk Assessments: The State of the Art

Objectives: Discussions of fairness in criminal justice risk assessments typically lack conceptual precision. Rhetoric too often substitutes for careful analysis. In this article, we seek to clarify

Causation in Antidiscrimination Law: Beyond Intent Versus Impact

Antidiscrimination law and scholarship for a long time has been engaged in the debate over whether a discriminatory intent or disparate impact test best captures the type of discrimination the law

A multidisciplinary survey on discrimination analysis

This survey is to provide a guidance and a glue for researchers and anti-discrimination data analysts on concepts, problems, application areas, datasets, methods, and approaches from a multidisciplinary perspective.

Counterfactual Fairness: Unidentification, Bound and Algorithm

This paper mathematically bounding the unidentifiable counterfactual quantity is addressed, and a theoretically sound algorithm for constructing counterfactually fair classifiers is developed.

The Sensitivity of Counterfactual Fairness to Unmeasured Confounding

This work designs tools to assess the sensitivity of fairness measures to this confounding for the popular class of non-linear additive noise models (ANMs), and gives a procedure for computing the maximum difference between two counterfactually fair predictors, where one has become biased due to confounding.

Eddie Murphy and the Dangers of Counterfactual Causal Thinking About Detecting Racial Discrimination

The model of discrimination animating some of the most common approaches to detecting discrimination in both law and social science—the counterfactual causal model—is wrong. In that model, racial

Experimental and Quasi-Experimental Designs

Measuring discrimination in algorithmic decision making

Various discrimination measures that have been used, analytically and computationally analyze their performance, and highlight implications of using one or another measure are reviewed to produce a unifying view of performance criteria when developing new algorithms for non-discriminatory predictive modeling.