Designing Fair, Efficient, and Interpretable Policies for Prioritizing Homeless Youth for Housing Resources

Azizi, Mohammad Javad; Vayanos, Phebe; Wilder, Bryan; Rice, Eric; Tambe, Milind

doi:10.1007/978-3-319-93031-2_3

Cited by 37 publications

(36 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, studies show that chronic patterns of homelessness affect a relatively small number of persons (33, 34). Homeless assistance continuously interacts with households at different stages of different trajectories, which makes accurate prediction of risk as well as response to interventions exceedingly difficult (5, 38, 44, 58, 95).…”

Section: Complexity In Causes and Responses To Homelessnessmentioning

confidence: 99%

See 1 more Smart Citation

Solving Homelessness from a Complex Systems Perspective: Insights for Prevention Responses

Fowler

Hovmand

Marçal

et al. 2019

Annu. Rev. Public Health

View full text Add to dashboard Cite

Homelessness represents an enduring public health threat facing communities across the developed world. Children, families, and marginalized adults face life course implications of housing insecurity, while communities struggle to address the extensive array of needs within heterogeneous homeless populations. Trends in homelessness remain stubbornly high despite policy initiatives to end homelessness. A complex systems perspective provides insights into the dynamics underlying coordinated responses to homelessness. A constant demand for housing assistance strains service delivery, while prevention efforts remain inconsistently implemented in most countries. Feedback processes challenge efficient service delivery. A system dynamics model tests assumptions of policy interventions for ending homelessness. Simulations suggest that prevention provides a leverage point within the system; small efficiencies in keeping people housed yield disproportionately large reductions in homelessness. A need exists for policies that ensure reliable delivery of coordinated prevention efforts. A complex systems approach identifies capacities and constraints for sustainably solving homelessness.

show abstract

Section: Complexity In Causes and Responses To Homelessnessmentioning

confidence: 99%

“…Other screening tools show similar challenges for targeting preventive services (13, 28, 44, 94). The difficulty in prediction reflects the complexity that underlies homelessness (5, 38, 58).…”

Section: Complexity In Causes and Responses To Homelessnessmentioning

confidence: 99%

Solving Homelessness from a Complex Systems Perspective: Insights for Prevention Responses

Fowler

Hovmand

Marçal

et al. 2019

Annu. Rev. Public Health

View full text Add to dashboard Cite

show abstract

“…Our approach thus enables the generalization of these MIP based models to general decision-making tasks in socially sensitive settings with diverse fairness requirements. Compared to the regression trees introduced in (Verwer and Zhang 2017), we consider more flexible decision tree models which allow for linear scoring rules to be used at each branch and at each leaf -we term these "linear branching" and "linear leafing" rules in the spirit of (Azizi et al 2018). Compared to (Bertsimas and Dunn 2017;Verwer and Zhang 2017) which require one hot encoding of categorical features, we treat branching on categorical features explicitly yielding a more interpretable and flexible tree.…”

Section: Related Workmentioning

confidence: 99%

Learning Optimal and Fair Decision Trees for Non-Discriminative Decision-Making

Aghaei

Azizi

Vayanos

2019

AAAI

Self Cite

113

124

View full text Add to dashboard Cite

In recent years, automated data-driven decision-making systems have enjoyed a tremendous success in a variety of fields (e.g., to make product recommendations, or to guide the production of entertainment). More recently, these algorithms are increasingly being used to assist socially sensitive decisionmaking (e.g., to decide who to admit into a degree program or to prioritize individuals for public housing). Yet, these automated tools may result in discriminative decision-making in the sense that they may treat individuals unfairly or unequally based on membership to a category or a minority, resulting in disparate treatment or disparate impact and violating both moral and ethical standards. This may happen when the training dataset is itself biased (e.g., if individuals belonging to a particular group have historically been discriminated upon). However, it may also happen when the training dataset is unbiased, if the errors made by the system affect individuals belonging to a category or minority differently (e.g., if misclassification rates for Blacks are higher than for Whites). In this paper, we unify the definitions of unfairness across classification and regression. We propose a versatile mixed-integer optimization framework for learning optimal and fair decision trees and variants thereof to prevent disparate treatment and/or disparate impact as appropriate. This translates to a flexible schema for designing fair and interpretable policies suitable for socially sensitive decision-making. We conduct extensive computational studies that show that our framework improves the state-of-the-art in the field (which typically relies on heuristics) to yield non-discriminative decisions at lower cost to overall accuracy.

show abstract

“…It is noteworthy that this stream of researches usually considers fairness at an individual level. Besides, several studies have also addressed the problem of scarce resource fair allocation such as organ transplantation (Alagoz et al, 2009;Bertsimas et al, 2013;Zou et al, 2020) and social services (Zardari et al, 2010;Azizi et al, 2018). These works usually construct allocation models to maximize certain objectives (e.g., overall life years from transplant) while maintaining fairness and produce the participants' rank ordering.…”

Section: Fairness Research In Operation Managementmentioning

confidence: 99%

Toward a Fairness-Aware Scoring System for Algorithmic Decision-Making

Yang¹,

Wu²,

Chang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Scoring systems, as simple classification models, have significant advantages in interpretability and transparency when making predictions. They facilitate humans' decision-making by allowing them to make a quick prediction by hand through adding and subtracting a few point scores and thus have been widely used in various fields such as medical diagnosis of Intensive Care Units. However, (un)fairness issues in these models have long been criticized, and the use of biased data in the construction of score systems heightens this concern. In this paper, we propose a general framework to create data-driven fairness-aware scoring systems. Our approach is first to develop a social welfare function that incorporates both efficiency and equity.Then, we translate the social welfare maximization problem in economics into the empirical risk minimization task of the machine learning community to derive a fairness-aware scoring system with the help of mixed integer programming. We show that the proposed framework provides practitioners or policymakers great flexibility to select their desired fairness requirements and also allows them to customize their own requirements by imposing various operational constraints. Experimental evidence on several real data sets verifies that the proposed scoring system can achieve the optimal welfare of stakeholders and balance the interpretability, fairness, and efficiency issues.

show abstract

Designing Fair, Efficient, and Interpretable Policies for Prioritizing Homeless Youth for Housing Resources

Cited by 37 publications

References 19 publications

Solving Homelessness from a Complex Systems Perspective: Insights for Prevention Responses

Solving Homelessness from a Complex Systems Perspective: Insights for Prevention Responses

Learning Optimal and Fair Decision Trees for Non-Discriminative Decision-Making

Toward a Fairness-Aware Scoring System for Algorithmic Decision-Making

Contact Info

Product

Resources

About