活动

Organizers: Shenzhen Research Institute of Big Data (深圳市大数据研究院 in Chinese)

                   Shenzhen International Center for Industrial and Applied Mathematics (深圳国际工业与应用数学中心 in Chinese)

Speaker: Dr. Timo Berthold

Language: English

Talk title: Machine Learning inside MIP solvers

Talk abstract: Modern MIP solvers consist of many subroutines that take care of different aspects of the solution process: presolving, cut generation, cut selection, primal heuristics, and so forth. For a given MIP, the solver has to make online decisions on which of multiple alternative instantiations of a subroutine to employ or how to combine them. While it is often hard to beat hand-crafted rules, the use of machine learning models for making those decisions has become more prominent in recent years. In this presentation, we will discuss four projects in which we used ML to improve the performance of the solvers Xpress and SCIP on general MIP benchmarks. Two topics relate to cutting planes, while the other two are concerned with numerical stability.

Bio:  Timo Berthold is a principal engineer at FICO and leading the MIP development team of the FICO Xpress MIP solver. Further, he is a private lecturer at the Mathematical Optimization department of TU Berlin. Timo Berthold is an expert on computational mixed-integer linear and nonlinear programming, on heuristic methods, and parallel computing. He has published over 50 research papers in this field, and his Ph.D. and Diploma thesis won multiple awards.

Time: 4:00 – 5:00 PM, Mar. 13^th, 2023 (Beijing time)

Zoom Meeting ID: 955 4774 7451 (Passcode: 328057)