About the speaker: Wenpin Tang is currently Tang Family Assistant Professor of Industrial Engineering and Operations Research, Columbia University. He works at the intersection of stochastic analysis, machine learning and quantitative finance. His primary research areas are continuous-time stochastic processes and probabilistic ranking models. Tang’s current research interest is to improve the efficiency of machine learning algorithms using stochastic tools, and to develop robust AI methodology for the emerging fintech market. Examples include diffusion generative modeling, analysis and design of blockchain mechanisms, and the curse of dimensionality in large interacting particle systems.

Abstract: Recently, there has been growing interest in guiding, or fine tuning pretrained diffusion models or LLMs for specific purposes, e.g., aesthetic quality of images, functional property of proteins, and downstream tasks in finance and operations management. In this talk, I will discuss several (principled) approaches, encompassing conditional guidance, regularization and reinforcement learning. Some applications will also be presented. The talk is based on a series of joint work with my students Haoxian Chen and Hanyang Zhao, as well as my colleagues David Yao, Henry Lam.