Haoran Yan

About

I am a Ph.D. student in Computational Science and Engineering at Georgia Tech, with Mathematics as my home unit. I am fortunate to be advised by Prof. Wenjing Liao. My research interests lie at the intersection of mathematical foundations of machine learning, scientific machine learning, and operator learning, with a focus on approximation and generalization theory, representation learning, and data-driven modeling of dynamical systems.

Before joining Georgia Tech, I received my B.S. degrees in Applied Mathematics and Statistics and Data Science from UC Santa Barbara, where I was advised by Prof. Paul J. Atzberger, Prof. Gareth W. Peters, and Prof. Mengyang Gu.

Publications & Preprints

Neural Scaling Laws of Transformer Operator NetworkPreprint

Haoran Yan, Zhongjie Shi, Yuanzhe Xi, Peng Chen, Wenjing Liao. arXiv:2609.33533, 2026.

Unsupervised Cell Segmentation by Fast Gaussian Processes

Laura Baracaldo*, Blythe King*, Haoran Yan*, Yizi Lin, Nina Miolane, Mengyang Gu. The New England Journal of Statistics in Data Science, 2026. *Co-first authors.

Optimal Multiple-Stopping Strategies for Dynamic Bond Ladder PortfoliosWorking paper

Mantana Chudtong, Gareth W. Peters, Avery Anderson, Haoran Yan. SSRN Working Paper, 2026.

From Position to Function: Exploring Word Distributions within Intonation Units in American English Conversation

Ryan Ka Yau Lai, Lu Liu, Haoran Yan, John W. Du Bois. Proceedings of SEMDIAL 2023 — MariLogue, 2023.

Selected Talks

Approximation and Generalization Guarantees of Transformers for Operator Learning between Function Spaces

Oral presentation, DeepMath 2026.

Understanding Denoising Autoencoders through the Manifold Hypothesis: A Geometric Perspective

Oral presentation, 4th Southeast Applied and Computational Math Student Workshop, Emory University, 2026.

Selected Posters

Neural Scaling Laws

Georgia Statistics Day 2026

Understanding Denoising Autoencoders through the Manifold Hypothesis poster

Denoising Autoencoders & Manifold Hypothesis

Georgia Scientific Computing Symposium 2026
USC–UQ Summer School 2026

$Predicting the Geometry of X-Ray Crystallography Diffraction Images Using Computer Vision poster$

X-Ray Crystallography with Computer Vision

UCSB Data Science Capstone Showcase 2024

Experience

Spring 2026

Graduate Research Assistant, Georgia Institute of Technology

School of Mathematics; advised by Prof. Wenjing Liao.

Fall 2024

NSF-Funded Research Assistant, UC Santa Barbara

Department of Mathematics; nonlinear dynamical systems and physics-informed machine learning.

Spring 2024

Undergraduate Researcher, SLAC National Accelerator Laboratory

Deep learning for X-ray crystallography image analysis.

Summer 2023

NSF-Funded Research Assistant, UC Santa Barbara

Computational Uncertainty Quantification Lab; cell segmentation and statistical learning.

Summer 2021

Data Analyst Intern, ByteDance, Beijing

Overseas Mathematics Education Department.

Academic Service

Chair / Co-organizer, SIAM Student Seminar

Georgia Institute of Technology, 2026–Present.

Organizing Team, 2nd Atlanta Mathematics of Data Science Bootcamp

Georgia Institute of Technology, 2026.