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.
more »
« less
To reduce or not to reduce: a study on spatio-temporal surveillance
- Award ID(s):
- 1650913
- PAR ID:
- 10171788
- Date Published:
- Journal Name:
- Environmental and Ecological Statistics
- Volume:
- 26
- Issue:
- 3
- ISSN:
- 1352-8505
- Page Range / eLocation ID:
- 217 to 238
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
ABSTRACT Bacteria associated with eukaryotic hosts can affect host fitness and trophic interactions between eukaryotes, but the extent to which bacteria influence the eukaryotic species interactions within trophic levels that modulate biodiversity and species coexistence is mostly unknown. Here, we used phytoplankton, which are a classic model for evaluating interactions between species, grown with and without associated bacteria to test whether the bacteria alter the strength and type of species interactions within a trophic level. We demonstrate that host-associated bacteria alter host growth rates and carrying capacity. This did not change the type but frequently changed the strength of host interspecific interactions by facilitating host growth in the presence of an established species. These findings indicate that microbiomes can regulate their host species’ interspecific interactions. As between-species interaction strength impacts their ability to coexist, our findings show that microbiomes have the potential to modulate eukaryotic species diversity and community composition. IMPORTANCE Description of the Earth’s microbiota has recently undergone a phenomenal expansion that has challenged basic assumptions in many areas of biology, including hominid evolution, human gastrointestinal and neurodevelopmental disorders, and plant adaptation to climate change. By using the classic model system of freshwater phytoplankton that has been drawn upon for numerous foundational theories in ecology, we show that microbiomes, by facilitating their host population, can also influence between-species interactions among their eukaryotic hosts. Between-species interactions, including competition for resources, has been a central tenet in the field of ecology because of its implications for the diversity and composition of communities and how this in turn shapes ecosystem functioning.more » « less
An official website of the United States government

