I am an incoming Applied Scientist on the Pricing Systems team at Amazon, where I will join in June 2026.

I am currently a Ph.D. student in the Department of Industrial Engineering and Management Sciences at Northwestern University. I have completed my Ph.D. in applied statistics and statistical learning, with a minor in simulation and stochastic analysis, and expect to graduate in June 2026. I was fortunate to be advised by Prof. Diego Klabjan.

I have also been fortunate to work with Prof. Imon Banerjee.

My research interests broadly span statistics, statistical machine learning, and applied probability. More specifically, my work focuses on theoretical and computational methods for uncertainty quantification in stochastic systems, including offline reinforcement learning, controlled Markov chains, and stochastic simulation models.

From June to September 2025, I interned with the Middle Mile Science team at Amazon as an Applied Scientist intern. My hosts were Phil Kaminsky and Adam Elmachtoub. From June to December 2024, I worked with the Global Pricing team at McDonald’s as a recommender systems researcher. My host was Zhivko Gochev.

Before joining Northwestern, I received my B.S. in Mathematics and Applied Mathematics from Fudan University and my M.S. in Mathematical Statistics from Purdue University, where I was fortunate to be advised by Prof. Raghu Pasupathy.

Please find my latest CV here.

My Erdős number is 3: Ziwei Su - Diego Klabjan - Craig A. Tovey - Paul Erdős.

Publications

Submitted Work

1. Model-based Bootstrap of Controlled Markov Chains
   Ziwei Su, Imon Banerjee, Diego Klabjan, 2026. Manuscript available upon request.

2. Central Limit Theorems for Transition Probabilities of Controlled Markov Chains [pdf]
   Ziwei Su, Imon Banerjee, Diego Klabjan, 2025. Submitted for publication.

Published Work

1. Differentiable Calibration of Inexact Stochastic Simulation Models via Kernel Score Minimization [pdf] [GitHub]
   Ziwei Su, Diego Klabjan
   International Conference on Artificial Intelligence and Statistics (AISTATS), 2025.

2. Overlapping Batch Confidence Intervals on Statistical Functionals Constructed from Time Series: Application to Quantiles, Optimization, and Estimation [pdf]
   Ziwei Su, Raghu Pasupathy, Yingchieh Yeh, Peter W. Glynn
   ACM Transactions on Modeling and Computer Simulation (ACM TOMACS), 2024.

3. A Modified Multinomial Baseline Logit Model with Logit Functions Having Different Covariates [pdf]
   Hao Ding, Ziwei Su, Xiaoqian Liu
   Communications in Statistics - Simulation and Computation, 2020.