Hengxu Lin

Hi, I'm Hengxu, a PhD candidate in Quantitative Marketing at Columbia Business School.

My research develops machine learning methods for decision-making in settings where preferences and data change across tasks and environments. I focus on personalization, with an emphasis on methods that are actionable, efficient, and grounded in theory.

A central theme of my work is understanding heterogeneity and distribution shift: how human preferences vary across people, evolve over time, and how a learning and decision-making system can remain lifelong effective. I draw on tools from statistics, causal inference, optimization, and large language models to design methods that adapt to dynamic, non-i.i.d. real-world data.

Before my doctoral studies, I worked in the Machine Learning Group at Microsoft Research Asia, where I built large-scale, real-time AI models for quantitative finance and contributed to collaborative projects with one of China's largest mutual fund companies.

I received an M.S. in Financial Economics from Columbia University, and a B.S. in Mathematics and B.A. in Accounting from Sun Yat-sen University.