Miaolan Xie

photo

I am an Assistant Professor in the Edwardson School of Industrial Engineering at Purdue University, with a courtesy appointment in the Department of Mathematics. My research lies at the intersection of mathematical optimization, stochastic processes, and machine learning, drawing on tools from optimization theory, statistics, and applied probability. I develop adaptive optimization algorithms that are both theoretically grounded and practically effective — enhancing the efficiency and reliability of existing methods, especially in challenging settings involving noisy or messy data.

I am actively seeking motivated PhD students to join my research group. Reach out if you'd like to work with me!

I completed my Ph.D. in Operations Research and Information Engineering at Cornell University, with Professor Katya Scheinberg. Prior to joining Purdue, I was a postdoctoral Principal Researcher at the University of Chicago Booth School of Business, affiliated with the Healthcare Initiative, working with Professor Dan Adelman. I obtained my Bachelor of Mathematics with majors in Pure Mathematics and Combinatorics and Optimization from the University of Waterloo. I completed my Master's degree with Levent Tunçel in Combinatorics and Optimization at the University of Waterloo. Before starting my Ph.D., I worked as a data scientist in Alibaba on the retail supply chain platform team, and prior to that I worked in Baidu and PwC Consulting. In the summer of 2022, I was a Givens Associate in the Mathematics and Computer Science Division at Argonne National Laboratory, working with Stefan Wild and Matt Menickelly.

Publications

* indicates authorship in alphabetical order by last name.












Talks

Reliable and Adaptive Stochastic Optimization in the Face of Messy Data (with Highly Corrupted Inputs and Heavy-Tailed noises)

Reliable Adaptive Stochastic Optimization for Messy Data: with High Probability Guarantees

High Probability Complexity Bounds for Adaptive Optimization Methods with Stochastic Oracles

Stochastic Adaptive Regularization Method with Cubics: A High Probability Complexity Bound

High Probability Iteration and Sample Complexity for Adaptive Step Search via Stochastic Oracles

High Probability Complexity Bounds for Line Search Based on Stochastic Oracles

High Probability Step Size Lower Bound for Adaptive Stochastic Optimization

ControlBurn: Feature Selection by Sparse Forests

Teaching & Service

Academic Service

Contact

Other