Abstract: Combinatorial optimization and global optimization are well-established areas in operations research and computer science. Until recently, its methods have focused on solving problem instances in isolation, ignoring that they often stem from related data distributions in practice. However, recent years have seen a surge of interest in using machine learning as a new approach for solving combinatorial problems, either directly as solvers or by enhancing exact solvers.

In this talk, we will give an overview of recent approaches for accelerating optimization algorithms using machine learning. Examples will include learning to branch with graph convolutional neural networks, learning to select convex relaxations in global optimization algorithms, learning to select neighborhood size in primal heuristics like local branching. We will also give a brief introduction of the software package Ecole for conducting such tasks.

Bio: Can Li obtained his bachelor’s degree from Tsinghua University, China, in Chemical Engineering. He completed his PhD in Chemical Engineering at Carnegie Mellon University. He joined the School of Chemical Engineering at Purdue University as an assistant professor in the fall of 2022. His research group is focused on optimization, machine learning, and applications in sustainable energy systems.