Zitao Song 宋子韬

About Me

I am a second-year Ph.D. student in Computer Science at Purdue University, where I am fortunate to be advised by Prof. David F. Gleich and supported by the Lynn Fellowship. I develop scalable and theoretically grounded optimization methods for large models and scientific discovery — I care as much about making training efficient in practice as about the mathematical structure that explains why it works.

Before Purdue, I was a Research Associate in the Agent Mediated Intelligence group at Nanyang Technological University, working with Prof. Bo An. I received my M.S. in Data Science from CUHK-Shenzhen, advised by Prof. Shuang Li, and my B.S. in Applied Mathematics from XJTLU, advised by Prof. Jia Meng and Prof. Jionglong Su.

• Optimization theory: matrix, nonsmooth & adaptive methods
• Stochastic Optimization
• Efficient LLM training & inference
• Reinforcement learning & GFlowNets