I am a fifth-year PhD student in mathematics at Yale University, fortunately advised by Prof. Yihong Wu. Prior to Yale, I was at the Courant Institute of Mathematical Sciences and mentored by Prof. Paul Bourgade.

Email: haoyu [dot] wang [at] yale [dot] edu

Research Interests

• Probability
• High-dimensional statistics
• Statistical learning

I am particularly interested in random matrix theory and graph learning/inference. Recently, I start my interest in large language models.

Publications

• Quantitative Universality for the Largest Eigenvalue of Sample Covariance Matrices
  Haoyu Wang
  The Annals of Applied Probability 34, no. 3 (2024).
• Random Graph Matching in Geometric Models: the Case of Complete Graphs
  Haoyu Wang, Yihong Wu, Jiaming Xu, Israel Yolou
  Conference on Learning Theory (COLT), 2022
• Automatic Dataset Construction (ADC): Sample Collection, Data Curation, and Beyond
  (Joint with Minghao Liu, Zonglin Di, Jiaheng Wei, et al)
  Under review at ICLR 2025, code available upon request
• Optimal Estimation of High-Dimensional Empirical-Bayes Regression
  Haoyu Wang, Yandi Shen, Yihong Wu
  working paper, available upon request, 2024
• Resampling Sensitivity of High-Dimensional PCA
  Haoyu Wang
  In submission, 2023+
• Optimal Smoothed Analysis and Quantitative Universality for the Smallest Singular Value of Random Matrices
  Haoyu Wang
  In submission, 2022+
• Learning Consensus via Majority Vote on Dynamic Community Networks
  Haoyu Wang, Jiaheng Wei, Zhenyuan Zhang
  In submission, 2022+
• A note on maximal operators for the Schrodinger equationon T1
  Yuqiu Fu, Kevin Ren, Haoyu Wang
  In submission, 2023+

Projects

• Social Network De-anonymization via GNN Graph Matching
  Summary: We proposed statistical and GNN-based machine learning algorithms to identify same users between different anonymized social networks based on graph topological information
  Language and Tools: Python, PyTorch, NumPy, SciPy