TY - GEN
T1 - Disaggregated Interventions to Reduce Inequality
AU - Bynum, Lucius
AU - Loftus, Joshua
AU - Stoyanovich, Julia
N1 - Publisher Copyright:
© 2021 ACM.
PY - 2021/10/5
Y1 - 2021/10/5
N2 - A significant body of research in the data sciences considers unfair discrimination against social categories such as race or gender that could occur or be amplified as a result of algorithmic decisions. Simultaneously, real-world disparities continue to exist, even before algorithmic decisions are made. In this work, we draw on insights from the social sciences brought into the realm of causal modeling and constrained optimization, and develop a novel algorithmic framework for tackling pre-existing real-world disparities. The purpose of our framework, which we call the "impact remediation framework,"is to measure real-world disparities and discover the optimal intervention policies that could help improve equity or access to opportunity for those who are underserved with respect to an outcome of interest. We develop a disaggregated approach to tackling pre-existing disparities that relaxes the typical set of assumptions required for the use of social categories in structural causal models. Our approach flexibly incorporates counterfactuals and is compatible with various ontological assumptions about the nature of social categories. We demonstrate impact remediation with a hypothetical case study and compare our disaggregated approach to an existing state-of-the-art approach, comparing its structure and resulting policy recommendations. In contrast to most work on optimal policy learning, we explore disparity reduction itself as an objective, explicitly focusing the power of algorithms on reducing inequality.
AB - A significant body of research in the data sciences considers unfair discrimination against social categories such as race or gender that could occur or be amplified as a result of algorithmic decisions. Simultaneously, real-world disparities continue to exist, even before algorithmic decisions are made. In this work, we draw on insights from the social sciences brought into the realm of causal modeling and constrained optimization, and develop a novel algorithmic framework for tackling pre-existing real-world disparities. The purpose of our framework, which we call the "impact remediation framework,"is to measure real-world disparities and discover the optimal intervention policies that could help improve equity or access to opportunity for those who are underserved with respect to an outcome of interest. We develop a disaggregated approach to tackling pre-existing disparities that relaxes the typical set of assumptions required for the use of social categories in structural causal models. Our approach flexibly incorporates counterfactuals and is compatible with various ontological assumptions about the nature of social categories. We demonstrate impact remediation with a hypothetical case study and compare our disaggregated approach to an existing state-of-the-art approach, comparing its structure and resulting policy recommendations. In contrast to most work on optimal policy learning, we explore disparity reduction itself as an objective, explicitly focusing the power of algorithms on reducing inequality.
KW - causal modeling
KW - fairness
KW - inequality
KW - social categories
UR - http://www.scopus.com/inward/record.url?scp=85119266808&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85119266808&partnerID=8YFLogxK
U2 - 10.1145/3465416.3483286
DO - 10.1145/3465416.3483286
M3 - Conference contribution
AN - SCOPUS:85119266808
T3 - ACM International Conference Proceeding Series
BT - Proceedings of 2021 ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization, EAAMO 2021
PB - Association for Computing Machinery
T2 - 2021 ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization, EAAMO 2021
Y2 - 5 October 2021 through 9 October 2021
ER -