Combating discrimination using Bayesian networks

By Koray Mancuhan, Chris Clifton. 2014

59 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Political Science and International Relations, Safety Research, Probabilistic & Bayesian Methods, AI Ethics, Fairness & Society

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 Political Science and International Relations

In Safety Research

In Probabilistic & Bayesian Methods

In AI Ethics, Fairness & Society

Citations in the corpus

Causal Modeling-Based Discrimination Discovery and Removal: Criteria, Bounds, and Algorithms (2019, neutral): “Based on that, [15] further proposed to use Bayesian networks to compute the confidence of the classification rules for detecting discrimination.”
A Causal Framework for Discovering and Removing Direct and Indirect Discrimination (2016, background): “Classiﬁcation rule-based meth-ods such as elift (Pedreshi, Ruggieri, and Turini 2008) and belift (Mancuhan and Clifton 2014) were proposed to represent certain discrimination patterns.”
Achieving Non-Discrimination in Data Release (2016, methodology): “Data preprocessing methods [1, 9, 12, 16, 23, 32, 34] modify the historic data to remove discriminatory effect before conducting predictive analysis.”
A survey on measuring indirect discrimination in machine learning (2015, methodology): “Data preprocessing may modify the target variable [Kamiran and Calders 2009; Mancuhan and Clifton 2014; Kamiran et al. 2013a], or modify input data [Feldman et al. 2015; Zemel et al. 2013].”
Fairness in Decision-Making - The Causal Explanation Formula (2018, background): “…fairness following, more or less explicitly, these two legal frameworks, including (not exhaustively) (Dwork et al. 2012; Romei and Ruggieri 2014; Mancuhan and Clifton 2014; Datta, Sen, and Zick 2016; Barocas and Selbst 2016; Hardt et al. 2016; Zhang, Wu, and Wu 2016; Kusner et al. 2017; Zafar…”
Exposing the probabilistic causal structure of discrimination (2015, background): “6 We could not compare with [26] due to repeatability issues.”
Using Loglinear Model for Discrimination Discovery and Prevention (2016, background): “The authors in [15] proposed the use of Bayesian networks to compute the confidence of the classification rules for discrimination discovery.”
Understanding Agent Incentives using Causal Influence Diagrams. Part I: Single Action Settings (2019, methodology): “A nuanced account of the situation can be obtained using causal graphs (Bonchi et al., 2017; Chiappa, 2019; Kilbertus et al., 2017; Kusner et al., 2017; Mancuhan and Clifton, 2014; Pearl, 2009; Zhang and Wu, 2017).”
Equality of Opportunity in Classification: A Causal Approach (2018, background): “Many attempts have been made to model discrimination in the classification setting [26, 14, 11, 9, 15].”
Building Better Open-Source Tools to Support Fairness in Automated Scoring (2017, background): “…complex non-linear algorithms: the training data can be modified (Feldman et al., 2015; Kamiran and Calders, 2012; Hajian and Domingo-Ferrer, 2013; Mancuhan and Clifton, 2014), the algorithm itself can be changed so that it optimizes for fairness as well as the selection criteria (Kamishima et…”
Anti-discrimination learning: a causal modeling-based framework (2017, methodology): “Based on that, methods for discrimination prevention have been proposed [1,7,9,12–15, 17– 19, 21, 35, This paper introduces a causal modeling-based framework for anti-discrimination learning.”
Algorithmic Bias: From Discrimination Discovery to Fairness-aware Data Mining (2016, methodology): “3 Data mining approaches for discrimination discovery: classification rule mining [16, 17, 18], K-NN classification [19], bayesian networks [20], probabilistic causation [21], privacy attack strategies [22].”
On Discrimination Discovery Using Causal Networks (2016, unknown): “Based on that, the authors in [12] further proposed to use the Bayesian network to compute the confidence of the classification rules for detecting discrimination.”
Situation Testing-Based Discrimination Discovery: A Causal Inference Approach (2016, methodology): “Based on that, [Mancuhan and Clifton, 2014] further proposed to use the Bayesian network to compute the confidence of the classification rules for detecting discrimination.”
Measuring discrimination in algorithmic decision making (2017, methodology): “The belift ratio (Mancuhan and Clifton 2014) is similar to the elift ratio in absolute measures, but here the probabilities of positive outcome are also conditioned on input attributes, where X = X r ∪ X r is a set of input variables, X r denotes so called redlining attributes, the variables which…”
Using sensitive personal data may be necessary for avoiding discrimination in data-driven decision models (2016, background): “…discrimination (Luong et al. 2011; Pedreschi et al. 2008; Hajian and Domingo-Ferrer 2013) and removing discrimination (Feldman et al. 2015; Mancuhan and Clifton 2014; Zemel et al. 2013; Kamiran et al. 2013, 2010; Calders et al. 2013; Kamishima et al. 2012), preventing such discrimination…”
Fairness in Machine Learning (2020, methodology): “Pre-Processing methods consist in transforming the training data to remove undesired biases [16, 21, 24, 28, 47, 51, 52, 53, 61, 64, 65, 66, 67, 86, 90, 91, 92, 112, 114, 116, 118, 121, 122, 127, 126, 164, 174, 188, 190, 191, 193].”