Duration: 52:19

1,130 views

13

Social and behavioral interventions are a critical tool for governments and communities to tackle deep-rooted societal challenges such as homelessness, disease, and poverty. However, real-world interventions are almost always plagued by limited resources and limited data, which creates a computational challenge: how can we use algorithmic techniques to enhance the targeting and delivery of social and behavioral interventions? This talk focuses on problems in combinatorial optimization and machine learning which underlie this question. I will discuss robust optimization, information gathering, and decision-focused learning as tools for decision making under uncertainty. Together, these techniques allow us to make decisions with limited data, strategically gather additional data when needed, and use this data to improve decisions by integrating combinatorial optimization problems into the training of machine learning models. I will illustrate these topics through a public health application, optimizing awareness campaigns for HIV prevention among homeless youth. Alongside social work collaborators, I have partnered with community organizations in Los Angeles to deploy algorithms for this problem, substantially increasing the intervention's impact compared to traditional approaches.

View slides and more at https://www.microsoft.com/en-us/research/video/algorithmic-social-intervention/