Fairness-Aware Neural Rényi Minimization for Continuous Features

Grari, Vincent; Lamprier, Sylvain; Detyniecki, Marcin

doi:10.24963/ijcai.2020/313

Cited by 16 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our future work, we will conduct theoretical analysis and empirical evaluation of density based fairness notions, e.g., SP and BGL, and notions for multiple sensitive attributes. Some recent work [13] proposed to use Hirschfeld-Gebelein-Rényi Maximum (HGR) correlation coefficient as a regression fairness notion to evaluate the independence between prediction and sensitive attributes. However, it is quite challenging to compute HGR.…”

Section: Discussionmentioning

confidence: 99%

Fair Regression under Sample Selection Bias

Du¹,

Wu²,

Tong³

2021

Preprint

View full text Add to dashboard Cite

Recent research on fair regression focused on developing new fairness notions and approximation methods as target variables and even the sensitive attribute are continuous in the regression setting. However, all previous fair regression research assumed the training data and testing data are drawn from the same distributions. This assumption is often violated in real world due to the sample selection bias between the training and testing data. In this paper, we develop a framework for fair regression under sample selection bias when dependent variable values of a set of samples from the training data are missing as a result of another hidden process. Our framework adopts the classic Heckman model for bias correction and the Lagrange duality to achieve fairness in regression based on a variety of fairness notions. Heckman model describes the sample selection process and uses a derived variable called the Inverse Mills Ratio (IMR) to correct sample selection bias. We use fairness inequality and equality constraints to describe a variety of fairness notions and apply the Lagrange duality theory to transform the primal problem into the dual convex optimization. For the two popular fairness notions, mean difference and mean squared error difference, we derive explicit formulas without iterative optimization, and for Pearson correlation, we derive its conditions of achieving strong duality. We conduct experiments on three real-world datasets and the experimental results demonstrate the approach's effectiveness in terms of both utility and fairness metrics.Preprint. Under review.

show abstract

Section: Discussionmentioning

confidence: 99%

Fair Regression under Sample Selection Bias

Du¹,

Wu²,

Tong³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Other works employ adversarial representation learning to remove protected factor information from latent representations, including [19]- [23]. Applications using this principle include [24] which uses adversarial learning to develop fair models of cardiovascular disease risk, while [25] explores the statistical properties of fair representation learning and [26] applies an adversarial approach for continuous features. Similar to our work, [27] employs an information-theoretic approach to learn fair representations.…”

Section: A Fairness Approachesmentioning

confidence: 99%

Achieving Utility, Fairness, and Compactness via Tunable Information Bottleneck Measures

Gronowski¹,

Paul²,

Alajaji³

et al. 2022

Preprint

View full text Add to dashboard Cite

Designing machine learning algorithms that are accurate yet fair, not discriminating based on any sensitive attribute, is of paramount importance for society to accept AI for critical applications. In this article, we propose a novel fair representation learning method termed the Rényi Fair Information Bottleneck Method (RFIB) which incorporates constraints for utility, fairness, and compactness of representation, and apply it to image classification. A key attribute of our approach is that we consider -in contrast to most prior work -both demographic parity and equalized odds as fairness constraints, allowing for a more nuanced satisfaction of both criteria. Leveraging a variational approach, we show that our objectives yield a loss function involving classical Information Bottleneck (IB) measures and establish an upper bound in terms of the Rényi divergence of order α on the mutual information IB term measuring compactness between the input and its encoded embedding. Experimenting on three different image datasets (EyePACS, CelebA, and FairFace), we study the influence of the α parameter as well as two other tunable IB parameters on achieving utility/fairness trade-off goals, and show that the α parameter gives an additional degree of freedom that can be used to control the compactness of the representation. We evaluate the performance of our method using various utility, fairness, and compound utility/fairness metrics, showing that RFIB outperforms current state-of-the-art approaches.

show abstract

“…We also assess the strength of nonlinear dependence between the signals and the claim counts with the help of the Hirschfeld-Gebelein-Renyi (HGR) maximal correlation coefficient proposed by Grari et al (2020). The HGR coefficient is equal to 0 if the two random variables are independent.…”

Section: Association Between Signals and Claim Countsmentioning

confidence: 99%

Joint Modeling of Claim Frequencies and Behavioral Signals in Motor Insurance

Corradin

Denuit²,

Detyniecki

et al. 2021

ASTIN Bull.

Self Cite

View full text Add to dashboard Cite

Telematicsdevices installed in insured vehicles provide actuaries with new risk factors, such as the time of the day, average speeds, and other driving habits. This paper extends the multivariate mixed model describing the joint dynamics of telematics data and claim frequencies proposed by Denuit et al. (2019a) by allowing for signals with various formats, not necessarily integer-valued, and by replacing the estimation procedure with the Expected Conditional Maximization algorithm. A numerical study performed on a database related to Pay-How-You-Drive, or PHYD motor insurance illustrates the relevance of the proposed approach for practice.

show abstract

Fairness-Aware Neural Rényi Minimization for Continuous Features

Cited by 16 publications

References 12 publications

Fair Regression under Sample Selection Bias

Fair Regression under Sample Selection Bias

Achieving Utility, Fairness, and Compactness via Tunable Information Bottleneck Measures

Joint Modeling of Claim Frequencies and Behavioral Signals in Motor Insurance

Contact Info

Product

Resources

About