<i>How fair can we go in machine learning</i>
            ? Assessing the boundaries of accuracy and fairness

Valdivia, Ana; Sánchez‐Monedero, Javier; Casillas, Jorge

doi:10.1002/int.22354

Cited by 38 publications

(18 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Backurs et al (2019), Chierichetti et al (2017), and Mahabadi andVakilian (2020), andHuang et al (2019). Feldman et al (2015), , Friedler et al (2019), Ignatiev et al (2020), Schelter et al (2020), Valdivia et al (2021.…”

Section: Dutch Census Datasetmentioning

confidence: 99%

A survey on datasets for fairness‐aware machine learning

Quy

Roy

Iosifidis

et al. 2022

WIREs Data Min & Knowl

View full text Add to dashboard Cite

As decision‐making increasingly relies on machine learning (ML) and (big) data, the issue of fairness in data‐driven artificial intelligence systems is receiving increasing attention from both research and industry. A large variety of fairness‐aware ML solutions have been proposed which involve fairness‐related interventions in the data, learning algorithms, and/or model outputs. However, a vital part of proposing new approaches is evaluating them empirically on benchmark datasets that represent realistic and diverse settings. Therefore, in this paper, we overview real‐world datasets used for fairness‐aware ML. We focus on tabular data as the most common data representation for fairness‐aware ML. We start our analysis by identifying relationships between the different attributes, particularly with respect to protected attributes and class attribute, using a Bayesian network. For a deeper understanding of bias in the datasets, we investigate interesting relationships using exploratory analysis. This article is categorized under: Commercial, Legal, and Ethical Issues > Fairness in Data Mining Fundamental Concepts of Data and Knowledge > Data Concepts Technologies > Data Preprocessing

show abstract

Section: Dutch Census Datasetmentioning

confidence: 99%

A survey on datasets for fairness‐aware machine learning

Quy

Roy

Iosifidis

et al. 2022

WIREs Data Min & Knowl

View full text Add to dashboard Cite

show abstract

“…Finally, this study advocates for evaluating face recognition algorithms along multiple axes of performance, namely overall effectiveness and fairness, using the Pareto curve method (Section 3.6). This has been proposed in other areas of ML fairness more broadly but this is the first study to specifically promote this technique and demonstrate its utility in the context of face recognition tasks [71], [72], [73].…”

Section: Pareto Curve Optimization With the Gini Aggregation Rate For...mentioning

confidence: 99%

Evaluating Proposed Fairness Models for Face Recognition Algorithms

Howard¹,

Laird²,

Sirotin³

et al. 2022

Preprint

View full text Add to dashboard Cite

The development of face recognition algorithms by academic and commercial organizations is growing rapidly due to the onset of deep learning and the widespread availability of training data. Though tests of face recognition algorithm performance indicate yearly performance gains, error rates for many of these systems differ based on the demographic composition of the test set. These "demographic differentials" in algorithm performance can contribute to unequal or unfair outcomes for certain groups of people, raising concerns with increased worldwide adoption of face recognition systems. Consequently, regulatory bodies in both the United States and Europe have proposed new rules requiring audits of biometric systems for "discriminatory impacts" (European Union Artificial Intelligence Act) and "fairness" (U.S. Federal Trade Commission). However, no international standard for measuring fairness in biometric systems yet exists. This paper characterizes two proposed measures of face recognition algorithm fairness (fairness measures) from scientists in the U.S. and Europe, using face recognition error rates disaggregated across race and gender from 126 commercial and open source face recognition algorithms. We find that both proposed methods have mathematical characteristics that make them challenging to interpret when applied to disaggregated face recognition error rates as they are commonly experienced in practice. To address this, we propose a set of interpretability criteria, termed the Functional Fairness Measure Criteria (FFMC), that outlines a set of properties desirable in a face recognition algorithm fairness measure. We further develop a new fairness measure, the Gini Aggregation Rate for Biometric Equitability (GARBE), and show how, in conjunction with the Pareto optimization, this measure can be used to select among alternative algorithms based on the accuracy/fairness trade-space. Finally, to facilitate the development of more robust and interpretable fairness measures in the face recognition domain, we have open-sourced our dataset of machine-readable, demographically disaggregated error rates. We believe this is currently the largest such dataset of its kind available to the ML fairness community.

show abstract

“…97% of all classifiers obtained a ratio of at least 10%. We conjecture that the difference between the two learning methods might be due to the local nature of tree methods, which may make them inherently less fair (see a related discussion in Valdivia et al, 2021).…”

Section: Comparing the Minimal Discrepancy To The Actual Discrepancymentioning

confidence: 99%

Inferring Unfairness and Error from Population Statistics in Binary and Multiclass Classification

Sabato¹,

Treister²,

Yom‐Tov³

2022

Preprint

View full text Add to dashboard Cite

We propose methods for making inferences on the fairness and accuracy of a given classifier, using only aggregate population statistics. This is necessary when it is impossible to obtain individual classification data, for instance when there is no access to the classifier or to a representative individual-level validation set. We study fairness with respect to the equalized odds criterion, which we generalize to multiclass classification. We propose a measure of unfairness with respect to this criterion, which quantifies the fraction of the population that is treated unfairly. We then show how inferences on the unfairness and error of a given classifier can be obtained using only aggregate label statistics such as the rate of prediction of each label in each sub-population, as well as the true rate of each label. We derive inference procedures for binary classifiers and for multiclass classifiers, for the case where confusion matrices in each subpopulation are known, and for the significantly more challenging case where they are unknown. We report experiments on data sets representing diverse applications, which demonstrate the effectiveness and the wide range of possible uses of the proposed methodology.

show abstract

How fair can we go in machine learning ? Assessing the boundaries of accuracy and fairness

Cited by 38 publications

References 19 publications

A survey on datasets for fairness‐aware machine learning

A survey on datasets for fairness‐aware machine learning

Evaluating Proposed Fairness Models for Face Recognition Algorithms

Inferring Unfairness and Error from Population Statistics in Binary and Multiclass Classification

Contact Info

Product

Resources

About