Fairness-aware Learning through Regularization Approach

By Toshihiro Kamishima, Shotaro Akaho, Jun Sakuma. 2011

464 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Safety Research, Optimization & Training, 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 Safety Research

In Optimization & Training

In AI Ethics, Fairness & Society

Citations in the corpus

Does Object Recognition Work for Everyone? (2019, background): “For instance, statistical parity [5, 6, 14, 23–25, 30, 49] states that in binary-classiﬁcation settings, members of different groups should have the same chance of receiving a positive prediction.”
Towards fairer datasets: filtering and balancing the distribution of the people subtree in the ImageNet hierarchy (2019, background): “…the data [39, 59], striking a tradeoff between performance and fairness by posing additional regularization during training or inference [34, 40, 52, 80, 81, 83], or designing application-specific interventions (such as of Burns et al. [11] for reducing gender bias in image captioning…”
Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods (2018, background): “For binary classiﬁcation Kamishima et al. (2012, 2011) add a regularization term to their objective that penalizes biased predictions.”
Certifying and Removing Disparate Impact (2014, methodology): “[9, 10] develop a regularizer for classifiers to penalize prejudicial outcomes and show that this can reduce indirect prejudice (their name for implicit discrimination like disparate impact) while still allowing for accurate classification.”
Inherent Trade-Offs in the Fair Determination of Risk Scores (2016, background): “Work in these directions has developed learning algorithms that penalize violations of statistical parity [4, 22].”
Advances and Open Problems in Federated Learning (2019, background): “ and conﬁrmation bias. One common antipattern is that minority or marginalized social groups are under-represented in the training data, and thus the learner weights these groups less during training [222], leading to inferior quality predictions for members of these groups (e.g. [85]). Just as the ”
Counterfactual Fairness (2017, background): “As a result, there has been immense interest in designing algorithms that make fair predictions (Bolukbasi et al., 2016; Calders & Verwer, 2010; Dwork et al., 2012; Grgic-Hlaca et al., 2016; Hardt et al., 2016; Joseph et al., 2016; Kamiran & Calders, 2009; 2012; Kamishima et al., 2011; Kleinberg et ”
Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions (2018, background): “learning processes (DLPs), which allow the use of sensitive variables during model building but not prediction, are either sub- or equi-optimal [102, 64, 66, 67, 131].”
Iterative Orthogonal Feature Projection for Diagnosing Bias in Black-Box Models (2016, methodology): “in (Kamishima et al., 2011) where they quantify prejudice by adding a mutual information based regularizer to the cost function of a logistic regression model.”
Efficient candidate screening under multiple tests and implications for fairness (2019, background): “In the more recent literature on fairness in machine learning, researchers often focus on binary classification, with employees characterized by a protected characteristic (group membership), and other (non-protected) covariates [15, 11, 12].”
Aequitas: A Bias and Fairness Audit Toolkit (2018, methodology): “…algorithmic fairness, and developing 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…”
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.”
FairGAN: Fairness-aware Generative Adversarial Networks (2018, background): “remove discriminatory effect before conducting predictive analytics [4], [7], [11]–[13], the in-process approaches that enforce fairness to classifiers by introducing constraints or regularization terms to the objective functions [14], [15], and the post-process approaches that directly change the p”
Adaptive Sensitive Reweighting to Mitigate Bias in Fairness-aware Classification (2018, background): “[29] categorize sources of unfair labeling as prejudice stemming from correlations between features and a sensitive attribute, underestimation due to inadequate convergence of the training algorithm and negative legacy of (historical) human biases in labeling training data.”
On Fairness and Calibration (2017, background): “However, it has been argued that Statistical Parity may not be applicable in many scenarios [7, 12, 18, 24], as it attempts to guarantee equal representation.”
Fairness in Criminal Justice Risk Assessments: The State of the Art (2017, methodology): “Instructive definitions can be found in computer science (Pedreschi, 2008; Kamishima et al., 2011; Dwork et al., 2012; Kamiran et al., 2012; Chouldechova, 2016; Friedler et al., 2016; Hardt et al., 2016; Joseph et al., 2016; Kleinberg et al., 2016; Calmon et al., 2017; Corbitt-Davies et al., 2017), ”