Shuotao Diao

Shuotao (Sonny) Diao

Assistant Professor
School of Advanced Engineering
Great Bay University, China

I am an Assistant Professor in the School of Advanced Engineering at Great Bay University. Prior to joining Great Bay University, I was a Postdoctoral Scholar in the Department of Industrial Engineering and Management Sciences at Northwestern University, working with Prof. David P. Morton.

I earned my Ph.D. in Industrial and Systems Engineering at the University of Southern California, under the supervision of Prof. Suvrajeet Sen. I also hold an M.S. in Computer Science (USC), an M.S. in Statistics (Lehigh University), and a B.S. in Physics (Shandong University).

sdiao@gbu.edu.cn sonnydiao430@gmail.com
Curriculum Vitae (PDF)

Research Interests

  • Stochastic programming and decomposition algorithms
  • Data-driven optimization under uncertainty
  • Approximate dynamic programming (multistage & infinite horizon)
  • Non-parametric statistical learning for optimization
  • Bilevel stochastic programming
  • Applications in healthcare, public health, and industrial systems

Selected Research Projects

Decomposition-Based Methods for Infinite-Horizon Multistage Stochastic Programs

2024 – Present

Study contraction and semicontraction, as well as fixed-point existence, for abstract dynamic programming mappings. Develop infinite-horizon variants of Stochastic Dual Dynamic Programming (SDDP) and Stochastic Decomposition (SD), integrated with compromise-decision strategies to improve solution reliability.

Efficient Solution Methods for Bilevel Stochastic Programming

2026 – Present

Investigate extreme-point enumeration, decomposition-based methods, and difference-of-convex approaches, as well as their possible fusions, for solving bilevel stochastic programs.

A Reliability Theory of Compromise Decisions for Large-Scale Stochastic Programs

2023 – 2025

Develop theoretical guarantees for compromise-decision variance-reduction strategies. Use Rademacher averages to derive sample-complexity bounds. Published in Mathematical Programming, 2026.

doi.org/10.1007/s10107-026-02336-2

Learning Enabled Optimization with Non-parametric Approximation

2016 – 2022

Integrate stochastic programming with k-NN and kernel estimation. Design a non-parametric stochastic quasi-gradient method for predictive stochastic programming with conditional-expectation objectives. Develop non-parametric extensions of Stochastic Decomposition. Published in Computational Optimization and Applications, 2024.

doi.org/10.1007/s10589-023-00529-5

Design of Staged Alert System with a Wastewater Signal

2023 – 2025

Incorporate daily viral-load measurements into an SEIR compartmental model to study the relationship between wastewater signals and hospital census. Design a wastewater-based staged-alert system to support public-health decision-making during epidemic surges. Manuscript in preparation.

Seminar Slides (PDF)

Pareto Optimal Graph Clustering

2023 – 2025

Use matroid theory and total dual integrality to characterize the pattern of Pareto optimal solutions to graph clustering problems. Design diversification strategies for column generation to improve the upper bound. Manuscript in preparation.

Publications

Peer-Reviewed Journals

  1. Diao, Shuotao, and Suvrajeet Sen. “A reliability theory of compromise decisions for large-scale stochastic programs.” Mathematical Programming. Preprint online, 2026. doi: 10.1007/s10107-026-02336-2
  2. Diao, Shuotao, and Suvrajeet Sen. “Distribution-free algorithms for predictive stochastic programming in the presence of streaming data.” Computational Optimization and Applications 87, no. 2 (2024): 355–395. doi: 10.1007/s10589-023-00529-5
  3. Eisenberg, Bennett, and Shuotao Diao. “Properties of the Kelly bets for pairs of binary wagers.” Statistics & Probability Letters 125 (2017): 215–219. doi: 10.1016/j.spl.2017.02.019

Peer-Reviewed Conferences

  1. Diao, Shuotao, and Suvrajeet Sen. “A unifying theory for the reliability of stochastic programming solutions using compromise decisions.” Proceedings of the INFORMS Optimization Society Conference, Houston, TX, March 2024.

Working Papers

  1. Shuotao Diao, David P. Morton, Oscar Dowson, and Bernardo Pagnoncelli. “A decomposition-based algorithm for infinite-horizon multistage stochastic programs on finite-state Markov chains.” Manuscript in preparation.
  2. Guyi Chen, Shuotao Diao, Katelyn Plaisier Leisman, David P. Morton, and Linda Pei. “Mitigating Pandemic Surges Using Wastewater Signals.” Manuscript in preparation.
  3. Shuotao Diao, Gökçe Kahvecioğlu, and David P. Morton. “Pareto Optimal Graph Clustering.” Manuscript in preparation.
  4. Diao, Shuotao, and Suvrajeet Sen. “Online Non-parametric Estimation for Nonconvex Stochastic Decomposition with Majorization-Minimization.” Optimization Online

Conference Presentations

  1. “A decomposition-based method for infinite horizon multistage stochastic programs on finite-state Markov chains” (with D. Morton, O. Dowson, B. Pagnoncelli). INFORMS Annual Meeting, San Francisco, CA, November 2026 (upcoming).
  2. “A reliability theory of compromise decisions for large-scale stochastic programs” (with S. Sen). INFORMS Annual Meeting, Seattle, WA, October 2024.
  3. “Comparisons between bagging and compromise decisions” (with S. Sen). INFORMS Annual Meeting, Phoenix, AZ, October 2023.
  4. “Non-parametric stochastic decomposition for predictive stochastic programming in the presence of streaming data” (with S. Sen). International Conference on Stochastic Programming XVI, Davis, CA, July 2023.
  5. “Non-parametric stochastic decomposition for two-stage predictive stochastic programming” (with S. Sen). INFORMS Annual Meeting, Anaheim, CA, October 2021.
  6. “Non-parametric stochastic quasi-gradient method in stochastic programming” (with S. Sen). INFORMS Annual Meeting, Seattle, WA, October 2019.
  7. “Learning enabled optimization with non-parametric estimation” (with S. Sen). International Conference on Stochastic Programming XV, Trondheim, Norway, August 2019.
  8. “Stochastic algorithms for conditional stochastic optimization” (with S. Sen). INFORMS Annual Meeting, Phoenix, AZ, November 2018.

Teaching

Instructor of Record Fall 2024

IEMS 351 – Optimization Methods in Data Science

Northwestern University, Department of Industrial Engineering and Management Sciences

  • Designed lectures, assignments, projects, and exams for an undergraduate optimization course.
  • Managed grading, office hours, course logistics, and supervision of one teaching assistant.
  • Course rating: 4.71 / 6.00 (vs. 4.18 under previous instructor, Fall 2023); Instructor rating: 5.00 / 6.00 (vs. 4.75, Fall 2023).
Guest Lecturer Spring 2022

ISE 638 – Stochastic Optimization (Graduate)

University of Southern California, Daniel J. Epstein Department of ISE

  • Delivered an 80-minute lecture on first-order stochastic programming algorithms and almost-sure convergence of stochastic approximation.
Guest Lecturer Fall 2019

ISE 330 – Introduction to Operations Research: Deterministic Models

University of Southern California, Daniel J. Epstein Department of ISE

  • Delivered four 80-minute lectures on linear algebra review, AMPL, the simplex algorithm, and Dijkstra's algorithm.
Teaching Assistant 2018 – 2022

University of Southern California

Daniel J. Epstein Department of Industrial and Systems Engineering

  • Undergraduate: ISE 225 Engineering Statistics I; ISE 220 Probability Concepts in Engineering; ISE 330 Introduction to Operations Research: Deterministic Models.
  • Graduate: ISE 630 Foundations of Optimization.

Contact & Service

Contact

Download CV

Professional Service

Referee for:

  • INFORMS Journal on Computing
  • Operations Research
  • INFORMS Journal on Optimization
  • SIAM Journal on Optimization
  • European Journal of Operational Research

Honors & Awards

  • Pearl-River Scholar (Youth), Guangdong Provincial Higher-Education Talent Program – 2025
  • Meritorious Reviewer, INFORMS Journal on Computing – 2025
  • Provost PhD Fellowship, USC – 2016
  • Outstanding Bachelor's Graduate, Shandong University – 2014

Software & Technical Skills

Languages: C++, Python, C, Java, Matlab, SQL, Xpress Mosel
Solver & Tools: CPLEX, Gurobi, AMPL, SciPy, NumPy, Pandas, scikit-learn, PyTorch, LaTeX