|
AI for Scientific Simulation and Discovery Lab, Westlake University
Research Intern 2025/08 – Present
Developing discrete diffusion models with improved efficiency and interpretability.
advised by Prof. Tailin Wu .
|
|
Michael Mahoney' Lab, UC Berkeley
undergraduate Research Assistant 2024/09 – 2025/08
Developed foundation models for scientific computing. Built SciML agents that can automatically generate code to solve scientific problems.
advised by Dr. Amir Gholami and Prof. Michael Mahoney.
|
|
AI for Scientific Simulation and Discovery Lab, Westlake University
Research Intern 2023/07 – 2024/06
AI for PDE/physical system simulation and control
advised by Prof. Tailin Wu .
|
|
Cyber-Med Laboratory, South China University of Technology
Student Research Project Leader 2022/04 – 2023/06
Biometric recognition in Human-Computper Interaction
advised by Prof. ZhanPeng
Jin and Prof. Gao Yang.
|
Education Experience
University of California, Berkeley, Berkeley, United States 08/24 – 05/25(expected)
GPA: 4/4
Main courses:
STAT 154/254 Modern Statistical Prediction and Machine Learning(A+)| CS 294 Physic-Inspired Machine
Learning(A)|
INFO 254 Applied Machine Learning(A)| CS 194/294 LLM Agent(A)| CS 194/294 Decentralized Finance(A).
South China University of Technology (SCUT), Guangzhou, China 09/21 – 06/25(expected)
B.Eng (Majoring in Artificial Intelligence)
GPA: 3.89/4 Rank: 3/80
Main courses:
Machine Learning| Deep Learning and Computer Vision| Artificial Intelligence and 3D Vision|
Signal and System| Calculus I/II| Linear Algebra| Probability and Statistics|
C++ Programming| Data Mining Data Structure &ensp
|
Contest
Mathematical Contest in Modeling(MCM/ICM), Meritorious Winner
InternationalGenetically Engineered Machine Competition(IGEM), Silver Award
Asia and Pacific Mathematical Contest in Modeling(APMCM), First Prize
|
Rewards
College Academic Innovation First-Class Scholarship
Hongping Changqing First-Class Scholarship
SCUT Second-Class Scholarship &ensp and so on, for a total of 7 scholarships.
|
|
Wavelet Diffusion Neural Operator
Peiyan Hu *,
Rui Wang *,
Xiang Zheng,
Tao Zhang,
Haodong Feng,
Ruiqi Feng,
Long Wei,
Yue Wang,
Zhi-Ming Ma,
Tailin Wu,
Wavelet Diffusion Neural Operator (WDNO) is a framework for PDE simulation and control. By modeling trajectories in the wavelet domain and using multi-resolution training, WDNO handles abrupt changes and generalizes across resolutions. It achieves state-of-the-art performance on five physical systems and a real-world dataset (ERA5), significantly improving long-term accuracy and reducing smoke leakage by 78% in a challenging 2D control task.
ICLR 2025
|
|
Updating
Author, updating
Updating
IMWUT 2024 2 revision/ 1 reject
|
|