AI4OPT Seminar Series
Date: Thursday, March 9, 2023
Time: Noon – 1:00 pm
Location: Instructional Center 115 (Scale Up Room) - (759 Ferst Dr, Atlanta, GA 30318)
Join Virtually: https://gatech.zoom.us/j/99381428980
Speaker: Garud Iyengar
Scalable Computation of Causal Bounds
Abstract: We consider the problem of computing bounds for causal inference problems with unobserved confounders, where identifiability does not hold. Existing non-parametric approaches for computing such bounds use linear programming~(LP) formulations that quickly become intractable for existing solvers because the size of the LP grows exponentially in the number of edges in the underlying causal graph. We show that this LP can be significantly pruned by carefully considering the structure of the causal query, allowing us to compute bounds for significantly larger causal inference problems. This pruning procedure also allows us to compute the bounds in closed form for a class of causal inference problems, which includes as a special case a well-studied family of problems where multiple confounded treatments influence an outcome. We extend our pruning methodology to fractional linear programs that are used to compute bounds for causal inference problems with observations. For causal inference without observations, we also propose a very efficient greedy heuristic that produces very high-quality bounds, and scales to problems that are several orders of magnitude larger than those for which the pruned LP can be solved.
This is joint work with Madhumitha Shridharan.
Link to conference paper: https://proceedings.mlr.press/v162/shridharan22a.html
Bio: Garud Iyengar is the Tang Family Professor of Operations, at Industrial Engineering and Operations Research at Columbia University, and the Senior Vice Dean for Research and Academic Programs at Columbia's College of Engineering. He received a BTech in electrical engineering from the Indian Institute of Technology Kanpur in 1993 and a PhD in electrical engineering from Stanford University in 1998. Prof. Iyengar is a member of Columbia’s Data Science Institute. His research is focused on understanding uncertain systems and exploiting available information using data-driven control and optimization algorithms. He and his students have explored applications in many diverse fields, such as machine learning, systemic risk, asset management, operations management, sports analytics, and biology. Iyengar was recognized as an INFORMS Fellow in 2018 and received an NSF CAREER award.
Lunch will be served at the seminar. So, please stop by 15 minutes before the seminar to pick up lunch.
To receive AI4OPT seminar announcements, please sign up to our mailing list. at https://lists.isye.gatech.edu/mailman/listinfo/ai4opt-seminars.
Past seminars can be found at https://www.ai4opt.org/seminars/past-seminars.