Selection in the Presence of Implicit Bias: The Advantage of Intersectional Constraints

Mehrotra, Anay; Pradelski, Bary; Vishnoi, Nisheeth K.

doi:10.1145/3531146.3533124

Cited by 8 publications

(17 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kleinberg and Raghavan [21] show that this type of bias naturally leads to underrepresentation of the disadvantaged group, and they study how a fairness mechanism called the Rooney rule affects the selection quality. This work is extended in [7,8,25]. Our work complements those studies by assuming strategic candidates who can respond to a policy chosen by a Bayesian decision-maker.…”

Section: Related Workmentioning

confidence: 95%

“…Recent literature on fairness in selection problem analyzed the problem using models based on two key ingredients to explain discrimination. Kleinberg and Raghavan [21] model selection problems with implicit bias (see also [7,8,25]), that is, where the decision-maker implicitly underevaluates the quality of the candidates from disadvantaged demographic groups. On the other hand, Emelianov et al [12,13] and Garg et al [16], following ideas from the economics literature on statistical discrimination (see Section 2), assume that the decision maker's estimate of the candidates quality is unbiased but has a higher variance for some demographic groups (a phenomenon terms implicit or differential variance).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fairness in Selection Problems with Strategic Candidates

Emelianov

Gast

Loiseau

2022

Proceedings of the 23rd ACM Conference on Economics and Computation

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 95%

Section: Introductionmentioning

confidence: 99%

Fairness in Selection Problems with Strategic Candidates

Emelianov

Gast

Loiseau

2022

Proceedings of the 23rd ACM Conference on Economics and Computation

View full text Add to dashboard Cite

“…Related work. Recent works [43,18,48] have studied the above model in the special case where the objective F is a linear function-a specific type of functions in F -and the bias functions, for each , are φ (x) = β • x for some parameters 0 < β 1 , β 2 , . .…”

Section: Introductionmentioning

confidence: 99%

“…There are many reasons to use fairness constraints, including, ethical and legal ones [74,61,62,6]. [43,18,48] demonstrate that another benefit of fairness constraints is that they can improve the latent utility of the output. Given u = (u 1 , u 2 , .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Maximizing Submodular Functions for Recommendation in the Presence of Biases

Mehrotra

Vishnoi

2023

Proceedings of the ACM Web Conference 2023

View full text Add to dashboard Cite

Subset selection tasks, arise in recommendation systems and search engines and ask to select a subset of items that maximize the value for the user. The values of subsets often display diminishing returns, and hence, submodular functions have been used to model them. If the inputs defining the submodular function are known, then existing algorithms can be used. In many applications, however, inputs have been observed to have social biases that reduce the utility of the output subset. Hence, interventions to improve the utility are desired. Prior works focus on maximizing linear functions-a special case of submodular functions-and show that fairness constraint-based interventions can not only ensure proportional representation but also achieve near-optimal utility in the presence of biases. We study the maximization of a family of submodular functions that capture functions arising in the aforementioned applications. Our first result is that, unlike linear functions, constraint-based interventions cannot guarantee any constant fraction of the optimal utility for this family of submodular functions. Our second result is an algorithm for submodular maximization. The algorithm provably outputs subsets that have near-optimal utility for this family under mild assumptions and that proportionally represent items from each group. In empirical evaluation, with both synthetic and real-world data, we observe that this algorithm improves the utility of the output subset for this family of submodular functions over baselines.

show abstract

Predictive policing and algorithmic fairness

Hung

Yen

2023

Synthese

View full text Add to dashboard Cite

This paper examines racial discrimination and algorithmic bias in predictive policing algorithms (PPAs), an emerging technology designed to predict threats and suggest solutions in law enforcement. We first describe what discrimination is in a case study of Chicago’s PPA. We then explain their causes with Broadbent’s contrastive model of causation and causal diagrams. Based on the cognitive science literature, we also explain why fairness is not an objective truth discoverable in laboratories but has context-sensitive social meanings that need to be negotiated through democratic processes. With the above analysis, we next predict why some recommendations given in the bias reduction literature are not as effective as expected. Unlike the cliché highlighting equal participation for all stakeholders in predictive policing, we emphasize power structures to avoid hermeneutical lacunae. Finally, we aim to control PPA discrimination by proposing a governance solution—a framework of a social safety net.

show abstract

Selection in the Presence of Implicit Bias: The Advantage of Intersectional Constraints

Cited by 8 publications

References 41 publications

Fairness in Selection Problems with Strategic Candidates

Fairness in Selection Problems with Strategic Candidates

Maximizing Submodular Functions for Recommendation in the Presence of Biases

Predictive policing and algorithmic fairness

Contact Info

Product

Resources

About