Decision and Control Laboratory
Coordinated Science Laboratory
“Optimization Over Probability Measures”
Yongxin Chen, Ph.D.
Georgia Institute of Technology
Wednesday, February 20, 2019
3:00pm – 4:00pm
CSL Auditorium (B02)
Optimization over probability measures is a class of optimization problems where the optimization variables are probability measures. Some typical examples include bayesian estimation, optimal mass transport and generative adversarial networks. In fact, any standard optimization can be reformulated into an optimization problem over probability measures. In this talk, I will cover both theories and algorithms for this type of problems. On the theory side, I will discuss its properties and connections to functional inequalities. On the algorithm side, I will cover some classical algorithms and compare them to a sample-based method we recently developed. Sever extensions such as minimax optimization will also be discussed if time allows.
Yongxin Chen received his B.S. in Mechanical Engineering from Shanghai Jiao Tong University, China, in 2011, and a Ph.D. degree in Mechanical Engineering from University of Minnesota in 2016. He currently serves as an Assistant Professor in the Daniel Guggenheim School of Aerospace Engineering at Georgia Institute of Technology. He has conducted researches in stochastic control, optimal transport and optimization. His current research focuses on the intersection between control, machine learning and optimization.