Hongseok Namkoong
Assistant Professor, Columbia University
namkoong@gsb.columbia.edu
I am an assistant professor in the Decision, Risk, and Operations division at Columbia Business School and a member of the Data Science Institute. My research interests lie at the interface of machine learning and decision-making. I develop robust and reliable methods for data-driven decision making, extending and connecting tools across ML, operations research, and causal inference. Outside of academia, I serve as a LinkedIn Scholar at LinkedIn’s Trust and Responsible AI team.
I received my Ph.D. from Stanford University and spent a year at Meta’s Adaptive Experimentation team as a research scientist. Here’s a more formal bio in the third person.
I go by Hong; alternatively, here’s a link the correct pronunciation of my first name.
news
Oct 15, 2024 | My Ph.D. student Ethan Che is on the faculty job market. He works on ML-driven approaches to large-scale operations problems and his Ph.D. thesis has taken strides in scalable auto-differentiation-based methods for queueing network control and adaptive experimentation. |
---|---|
Jun 10, 2024 | In this significantly updated paper, we use rigorous and extensive empirical analysis to highlight the shortcomings of robust learning algorithms and advocate for new data-driven inductive paradigm for algorithm development. Also check out our NeurIPS 2023 tutorial on distribution shifts and my DRO brown bag overviewing my group’s work in this direction. |
Dec 15, 2023 | Watch my SNAPP talk on Adaptive Experimentation at Scale. |
May 15, 2023 | Watch my talk on Robust Causal Inference that I gave at the IFDS workshop on distributional robustness in data science. |
selected publications
-
AI-driven DecisionsarXiv:2409.03740 [cs.LG], 2024
-
AI-driven DecisionsarXiv:2408.04570 [cs.LG], 2024Selected for oral presentations at the Econometric Society Interdisciplinary Frontiers: Economics and AI+ML conference and Conference on Digital Experimentation
-
AI-driven DecisionsarXiv:2405.19466 [cs.LG], 2024Selected for presentation at the Econometric Society Interdisciplinary Frontiers: Economics and AI+ML conference