Dr. Yiping Lu

Biography

Yiping Lu is an Assistant Professor in the Department of Industrial Engineering and Management Sciences at Northwestern University. He obtained his Ph.D. in Applied Computational Mathematics from Stanford University, and his B.S. in Computational Mathematics from Peking University. His research primarily revolves around scaling laws in machine learning and aims to understand the scaling behavior of machine learning systems, leading to improvements in performance and resource allocation. Professor Lu's work explores crucial aspects such as the interaction between approximation error, optimization difficulty, and statistical uncertainty as models scale up in size and complexity. This includes developing predictive scaling theories and new optimization methods to ensure stable performance as hyperparameters and model sizes increase. His interests expand to inference-time scaling as well, where he devises techniques to improve model accuracy through optimized computation. His publications cover a range of topics from diffusion models and neural optimizers to kernel operator learning and deep learning applications, highlighting his contributions to understanding the intricacies of machine learning methodologies. Professor Lu is committed to advancing the field and has presented his findings at international conferences, positioning himself as a key researcher in the integrative strategies of scalable machine learning.

Research Interests

Experience