Behavioral Policy Speaker Series: Combining Behavioral Science and Machine Learning to Develop Personalized Policies

Thu, Feb 22, 2018, 4:30 pm
Robertson Hall, Bowl 2

Jens Ludwig, the McCormick Foundation Professor of Social Service Administration, Law, and Public Policy at the University of Chicago Harris School of Public Policy, will deliver a talk entitled “Personalized Policies.” New insights from behavioral science have led to a wide range of policy interventions that were unheard of even just a few decades ago: text message reminders, switching defaults, or other changes in the choice environment like changing the height of the fruit versus candy in the supermarket check-out line. These interventions have been enormously successful in “nudging” people in situations where there is an obviously right thing to do. But for many key decisions, what the right thing to do is will vary across people. For example, consequential decisions like how much house to buy, who to hire, or which defendant a judge should grant bail all hinge on an individualized prediction; the right answer is “it depends.” We know that these tasks will be enormously difficult for people to get right, yet standard one-size-fits-all approaches may be of limited help. In this talk, Ludwig, will discuss how progress on these problems is possible by combining new tools from the computer science field of machine learning with insights from social science and behavioral science to develop new personalized policies.