Fairness Constraints: Mechanisms for Fair Classification

By Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez-Rodriguez, Krishna P. Gummadi. 2015

1,360 citations recorded by the source corpus.

Subfields: Safety, Robustness & Interpretability

How researchers see this paper

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

In Safety, Robustness & Interpretability

Citations in the corpus

FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age (2019, background): “Studies in algorithmic fairness have focused on either 1) discovering (auditing) existing bias in datasets or systems [50, 7, 30], 2) making a better dataset [39, 1], or 3) designing a better algorithm or model [12, 1, 47, 71, 70].”
Towards fairer datasets: filtering and balancing the distribution of the people subtree in the ImageNet hierarchy (2019, background): “Finally, learning with constraints is a di cult open problem, frequently resulting in satisfying fairness constraints at the expense of overall model accuracy [39, 40, 79, 80].”
A Survey on Bias and Fairness in Machine Learning (2019, methodology): “Domain Sub-domain Reference(s) Data Simpson’s Paradox [68] Machine learning Classification [65] [86] [49] [72] [120] [55] [130] [126] [59] [24] [127] Machine learning Regression [13] [1] Machine learning PCA [111] Machine learning Community detection [85] Machine learning Clustering [30] [7] Machine”
Fair Resource Allocation in Federated Learning (2019, background): “Another set of works optimize an objective subject to some fairness constraints during training time [3, 6, 18, 43, 46, 47, 9].”
Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions (2018, background): “Therefore, solutions such as disparate learning processes (DLPs) , which allow the use of sensitive variables during training but not prediction, are either sub-or equi-optimal [104, 67, 68, 69, 128].”
Contamination Attacks and Mitigation in Multi-Party Machine Learning (2018, background): “Preventing contamination attacks can be seen as ensuring fairness [9, 44, 45] from the trained models w.r.t. the contaminated attributes.”
A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual &Group Unfairness via Inequality Indices (2018, background): “Finally, designers of fair learning models make a number of ad hoc or implicit choices about fairness conditions without explicit justification; for instance, it is unclear why in many of the previous works [12, 17, 38, 39], the relative sizes of the groups in the population are not taken into accou”
Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning (2018, background): “their lack of fairness towards minorities [94], [95], [96], [97],”
Attributing Fair Decisions with Attention Interventions (2021, background): “…(Gupta et al. 2021; Moyer et al. 2018; Roy and Boddeti 2019; Jaiswal et al. 2020; Song et al. 2019), adding regularizers to incorporate fairness (Zafar et al. 2017; Kamishima et al. 2012; Mehrabi, Huang, and Morstatter 2020), or modifying the outcomes of unfair machine learning algorithms to…”
Statistical Equity: A Fairness Classification Objective (2020, methodology): “Some other methods perform in-processing which tries to incorporate the fairness objective during training phase [Kamishima et al., 2012, Zafar et al., 2015, Wu, Zhang, and Wu, 2018].”
Causal Modeling-Based Discrimination Discovery and Removal: Criteria, Bounds, and Algorithms (2019, background): “In [21], the authors added the measure of fairness into the classification learn-”
Stable and Fair Classification (2019, methodology): “They observed that the standard deviation of the fairness metric is 2.4 % for the algorithm in [41] ( KAAS ) with respect to the race attribute, and is 4.1 % for that in [75] ( ZVRG ) with respect to the sex attribute.”
Fairness Under Unawareness: Assessing Disparity When Protected Class Is Unobserved (2018, neutral): “Demographic disparity is simple to understand and is widely used [25, 30, 32], despite its flaws [17, 21].”
Aequitas: A Bias and Fairness Audit Toolkit (2018, methodology): “…methods to avoid bias and disparate impact or defining tradeoffs among different criteria (Calders and Verwer, 2010; Kamishima et al., 2011; Dwork et al., 2012; Zemel et al., 2013; Feldman et al., 2015; Hardt et al., 2016; Kleinberg et al., 2016; Corbett-Davies et al., 2017; Zafar et al., 2017b,a).”
Taking Advantage of Multitask Learning for Fair Classification (2018, background): “In recent years there has been a lot of interest in the problem of enhancing learning methods with “fairness” requirements, see [1– 4 8–10 12 17 19 20 22–26 28 30 31 35–39] and references therein.”