Portrait of Zihan Xu

Zihan Xu

Ph.D. Candidate in Applied Mathematics
University of California, Merced

Seeking postdoctoral or research scientist opportunities in mathematical oncology.

I develop Statistics-Informed Neural Networks (SINNs), recurrent neural-network surrogates that learn and generate stochastic time trajectories by matching ensemble statistics, without requiring trajectory-wise correspondence. My research interests span stochastic modeling, scientific machine learning, data-driven surrogate modeling, multiscale and hybrid simulation, and uncertainty quantification, with growing interests in bioinformatics and mathematical oncology.

Advisor: Prof. Changho Kim · Expected graduation: May 2027

News

Research

Statistics-informed neural networks for multidimensional stochastic processes

Developing recurrent neural-network surrogates trained through ensemble statistics. Incorporating marginal and cross-variable statistics extends SINN to multidimensional stochastic processes without requiring trajectory-wise correspondence.

Graduate Student Researcher, UC Merced · Fall 2022–present. First-author manuscript under review, 2026.

SINN architecture comparison

Conducting a comparison of LSTM, GRU, Transformer, and Mamba architectures for Statistics-Informed Neural Networks. Manuscript in preparation.

Presented at SIAM NCC26, UC Davis, October 2026; accepted for SIAM MDS26, Salt Lake City, November 2026. Poster (NCC26 / MDS26, PDF)

Stochastic dynamical systems with a slowly varying parameter

Extending SINN to stochastic systems with slowly varying parameters using a branch–trunk recurrent architecture to reproduce parameter-dependent stochastic trajectories.

Ongoing work · Invited CSE27 talk.

Hybrid simulation at Lawrence Berkeley National Laboratory

Embedded SINN as a neural surrogate for the fast stochastic component of a two-way coupled particle–continuum hybrid simulation. Worked on integrating generated stochastic trajectories into the multiscale workflow for computationally efficient simulation.

Research Intern, Applied Mathematics and Computational Research Division · Summer 2025. Mentor: Dr. Andy Nonaka.

Collaborative research projects

Contributed to DOE's Machine-Learning-Based Surrogate Modeling for Stochastic Multiscale Simulation Methodology (DE-SC0024240) and NSF's LEAPS-MPS: Stochastic Particle-Continuum Hybrid Simulation Method for Model Heterogeneous Catalysts under Reaction Conditions (CHE-2213368).

Manuscript Under Review

  1. Learning and Generating Time Trajectories of Multidimensional Stochastic Processes.
    Zihan Xu, Changho Kim, Andy J. Nonaka, and Yuanran Zhu.
    Under review, 2026. arXiv:2611.17711.

Manuscript in Preparation

  1. Comparing LSTM, GRU, Transformer, and Mamba for Statistics-Informed Neural Networks.
    Zihan Xu, Changho Kim, Roummel F. Marcia, Andy J. Nonaka, and Yuanran Zhu.
    Manuscript in preparation. Poster (NCC26 / MDS26, PDF).

Selected Presentations

Education & Training

University of California, Merced

Ph.D. in Applied Mathematics · Aug. 2021–expected May 2027 · GPA: 4.00/4.00

Selected coursework: Scientific Computing and Deep Learning.

University of Southern California

M.S. in Mathematical Finance · Aug. 2017–May 2019 · GPA: 3.87/4.00

Iowa State University

B.S. in Applied Mathematics: Actuarial Science · Aug. 2010–Dec. 2014 · GPA: 3.73/4.00

Applied Bioinformatics, UC San Diego Extended Studies

Specialized Certificate in Applied Bioinformatics, expected 2026.

Completed: Bioinformatics Fundamentals (BIOL-40236; A) and Genomic Sequencing Technologies (BIOL-40237; A+).

In progress: Processing Actionable Data in Genomics (BIOL-40327) and Modern-Omics: Technologies and Applications (BIOL-40374).

Harvard Medical School HMX

Cancer Genomics and Precision Oncology: The Genetic Basis of Cancer and Its Treatment.

Completed · July–October 2026.

Selected Honors & Awards

Teaching

Teaching Assistant · University of California, Merced

  • Calculus II (MATH 012), Fall 2026
  • Probability and Statistics (MATH 032), Fall 2025
  • Linear Analysis I (MATH 141), Spring 2023
  • Calculus I (MATH 021), Fall 2022
  • Linear Algebra and Differential Equations (MATH 024), Fall 2021, Spring 2022, and Summer 2022

Pedagogical Training and Certificates

Technical Skills

Programming

Python, C++, MATLAB, Mathematica, R, SQL, and Linux shell scripting.

High-performance computing

MPI and the Pinnacles HPC cluster at UC Merced.

Scientific computing & machine learning

PyTorch, NumPy, pandas, SciPy, scikit-learn, Matplotlib, Seaborn, numerical methods, and recurrent neural networks.

Tools & environments

Git, Docker, Jupyter, VS Code, PyCharm, and LaTeX.