CV

Blog

💬 Email

📖 Google Scholar

🧑‍💻 Github

🔍 OpenReview

About Me

I am now a Ph.D. candidate in the Data Science and Analytics Thrust at Hong Kong University of Science and Technology (Guangzhou), working on neural solvers for Partial Differential Equations and on accelerating scientific and industrial simulations.

I am supervised by Prof. Wei WANG and co-supervised by Prof. Zhilu LAI.

My current research interests include:

• Theoretical and structural properties of tensors in Neural Operators, and their use in applications
• Uncertainty Quantification for scientific machine learning and industrial simulation
• Closed-loop neural methods that support efficient production workflows

I am looking for research collaborations and industry internships. Feel free to write to me at [email protected], ideally with a short proposal or a brief outline of the problem your team wants to solve.

Education

2023 ~ Present: Ph.D. Student in Data Science and Analytics

• The Hong Kong University of Science and Technology (GZ), Guangzhou

2019 ~ 2023: B.Eng. in Mechatronics

• Tongji University, Shanghai

News

• We are building an open repository for parameterizing functionals in scientific machine learning within function spaces. If you are interested in participating, please contact me.

• Recently, we have submitted our work on epistemic uncertainty quantification for neural operators, introducing a novel scheme that reframes weight-space epistemic uncertainty as noise perturbations confined to the lifting-weight subspace (Efficient Monte-Carlo).

• We have published our paper: Reformulating Neural Operators in $d+1$ Dimensions for Embedding Evolution, which provides the first dual-basis view on embedding-space evolution in neural operators (Aggregations Also Benefit Embedding).

• We have published our paper: Harnessing scale and physics: A multi-graph neural operator framework for PDEs on arbitrary geometries, which adaptively constructs multi-scale graphs to support information aggregation in a graph-former architecture (Unifying Representations for NOs).

Skills

• Programming: Python (machine learning), C/C++, MATLAB (scientific computing)
• Languages: Mandarin Chinese (native), English (academic working proficiency), German (B2)
• Others:
  • Red-Hat
  • CC/Grok Advances
  • Bartending

Ramblings

Modified from the Academic Page template created by @Jiazan Wang