معرفی
Zhongyuan Lyu is a Research Fellow (Postdoctoral Research Scientist) at Columbia University's Data Science Institute, mentored by Professors Yuqi Gu and Kaizheng Wang. His research focuses on statistical methodology for latent structures in mixture models, graphical models, and tensor decompositions, with applications to heterogeneous data analysis. Prior to Columbia, he earned his PhD in Mathematics from the Hong Kong University of Science and Technology under Professor Dong Xia's supervision.
His academic background includes advanced work in high-dimensional data analysis, latent variable modeling, and computational statistics. Research interests emphasize developing theoretically grounded algorithms for complex data types, particularly in network science and multilayer data frameworks.
Recent publications highlight contributions to spectral clustering optimization, adaptive transfer learning frameworks, and tensor-based methodologies for higher-order networks. His work bridges statistical theory and practical applications, addressing computational limits and optimal estimation challenges in modern data science problems.
No scientific awards or grants are explicitly mentioned in the provided data. His current position is full-time within the Data Science Institute's research team.
