ORIE Colloquium: Kuang Xu (Standford), April 13, 2021

“Using mean-field modeling to power experimentation”

I will discuss a new approach to experimental design in large-scale stochastic systems with considerable cross-unit interference, under an assumption that the interference is structured enough that it can be captured via mean-field modeling. While classical approaches to experimental design assume that intervening on one unit does not affect other units,  there are many important settings where this noninterference assumption does not hold, such as when running experiments on supply-side incentives on a ride-sharing platform or subsidies in an energy marketplace. Our approach enables us to accurately estimate the effect of small changes to system parameters by combining unobtrusive randomization with lightweight modeling, all while remaining in equilibrium. We can then use these estimates to optimize the system by gradient descent. Concretely, we focus on the problem of a platform that seeks to optimize supply-side payments p in a centralized marketplace where different suppliers interact via their effects on the overall supply-demand equilibrium, and show that our approach enables the platform to optimize p in large systems using vanishingly small perturbations.

Paper: https://doi.org/10.1287/mnsc.2020.3844

Kuang Xu is an associate professor in the area of Operations, Information and Technology at Stanford Graduate School of Business, and an associate professor by courtesy with the Electrical Engineering Department, Stanford University. Born in Suzhou, China, he received the B.S. degree in Electrical Engineering (2009) from the University of Illinois at Urbana-Champaign, and the Ph.D. degree in Electrical Engineering and Computer Science (2014) from the Massachusetts Institute of Technology. His research primarily focuses on understanding fundamental properties and design principles of large-scale stochastic systems using tools from probability theory and optimization, with applications in queueing networks, healthcare, privacy and machine learning. He received First Place in the INFORMS George E. Nicholson Student Paper Competition (2011), the Best Paper Award, as well as the Kenneth C. Sevcik Outstanding Student Paper Award at ACM SIGMETRICS (2013), and the ACM SIGMETRICS Rising Star Research Award (2020). He currently serves as an Associate Editor for Operations Research.