Controlling Attribute Effect in Linear Regression

By Toon Calders, Asim Karim, Faisal Kamiran, Wasif Ali, Xiangliang Zhang. 2013

159 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability, Causal Inference & Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

In Causal Inference & Discovery

Citations in the corpus

Conscientious Classification: A Data Scientist's Guide to Discrimination-Aware Classification (2017, neutral): “16 Note the similarity between the conditional mean difference and the causal mean difference deﬁned earlier.”
Fair Regression: Quantitative Definitions and Reduction-based Algorithms (2019, background): “Calders et al. (2013), Johnson et al. (2016) and Komiyama et al. (2018) consider a more limited form of independence, expressed via a small number of moment constraints, such as lack of correlation, and design speciﬁc algorithms for linear least squares.”
Adaptive Sensitive Reweighting to Mitigate Bias in Fairness-aware Classification (2018, background): “xi,yi=1,i<S ∼ N ([1, 2], [5, 2; 2, 5]) xi,yi=1,i ∈S ∼ N ([2, 0], [10, 1; 1, 4]) xi,yi=0,i<S ∼ N ([0,−1], [7, 1; 1, 7]) xi,yi=0,i ∈S ∼ N ([−5, 0], [5, 1; 1, 5])”
A Convex Framework for Fair Regression (2017, background): “[6] introduced the study of statistical parity’s analog in regression settings, (called equal means and balanced residuals).”
Learning Fair Classifiers (2015, background): “The second strategy consists of modifying existing classifiers to limit discrimination (Cadlers & Verwer, 2010; Calders et al., 2013; Kamiran et al., 2012; Kamishima et al., 2011; 2013; Pedreschi et al., 2009).”
Causal Discrimination Discovery Through Propensity Score Analysis (2016, methodology): “The discrimination-aware data mining community has recently recognized the importance of causal analysis and the perils of correlation analysis [3, 6] while causation has been recognized as significant in the legal circles for a while [10].”
Fair Forests: Regularized Tree Induction to Minimize Model Bias (2017, methodology): “Others have taken to creating a differentiable regularization term, and applying it to models such as Logistic and Linear Regression [1, 2, 6, 18].”
Fairness in Machine Learning: A Survey (2020, background): “Reference [51] considers biases in linear regression as measured by the effects of a sensitive attribute on Y through the mean difference (difference of mean target variable between groups) and AUC metrics.”
Fairness in Machine Learning (2020, methodology): “Some methods transform the constrained optimization problem via the method of Lagrange multipliers [3, 15, 34, 37, 38, 59, 97, 135, 183, 185] or add penalties to the objective [5, 14, 22, 44, 46, 56, 58, 62, 73, 74, 75, 79, 82, 87, 88, 89, 93, 94, 95, 98, 104, 115, 119, 123, 133, 134, 135, 138, 139,”
Nonconvex Optimization for Regression with Fairness Constraints (2018, background): “We also show RMSE as a function of the mean difference (MD) and the area under the curve (AUC) (Calders et al., 2013) in Appendix D, which behaved similarly to the correlation coefficients.”
Impartial Predictive Modeling: Ensuring Fairness in Arbitrary Models (2016, methodology): “Others have argued for penalizing discrimination during estimation [8, 9, 17], modifying the raw data [11, 21], or modifying objective functions with fairness criteria [12, 13, 15, 25].”
Measuring discrimination in algorithmic decision making (2017, methodology): “It has been used by Calders et al. (2013) for measuring discrimination between two groups when the target variable is numeric (regression task), where I ( true ) = 1 and 0 otherwise.”
What About Applied Fairness? (2018, background): “…problems in ML; let alone nascent deﬁnitions in sub-areas like recommender systems (Burke et al., 2018; Ekstrand et al., 2018), regression (Calders et al., 2013; Berk et al., 2017), and clustering (Chierichetti et al., 2017) or those that may have been de-ﬁned in neighboring ﬁelds like…”
New Fairness Metrics for Recommendation that Embrace Differences (2017, background): “[2] propose to impose constrains on the residuals of linear regression models, which requires not only the mean prediction but also the mean residuals to be the same across groups.”
Discrimination-aware data mining: a survey (2017, methodology): “Recently, Calders et al. (2013) have provided a solution to discrimination-aware regression problem.”
Fairness Behind a Veil of Ignorance: A Welfare Analysis for Automated Decision Making (2018, methodology): “• Positive residual difference [Calders et al., 2013] is computed by taking the absolute difference of mean positive residuals across groups: ∣∣∣∣∣ 1 |G1 | ∑ i∈G1 max{0, (ŷi − yi)} − 1 |G2 | ∑ i∈G2 max{0, (ŷi − yi)} ∣∣∣∣∣ .”
Ethical Machine Learning in Healthcare (2020, background): “Regressions subject to group fairness constraints or penalties optimizing toward joint global and group fit considerations have also been developed (118, 119, 93).”
Fairness Constraints: Mechanisms for Fair Classification (2015, background): “ave the task of specifying the exact value of the ratio to the system administrator. The second strategy consists of modifying existing classiﬁers to limit discrimination (Cadlers & Verwer, 2010; Calders et al., 2013; Kamiran et al., 2012; Kamishima et al., 2011; 2013; Pedreschi et al., 2009). A”