科研活动
“Mathematical Programming Computation” Seminar Series: Prof. Thorsten Koch
“Mathematical Programming Computation” Seminar Series: Prof. Thorsten Koch
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: Prof. Thorsten Koch
Language: English
Talk title: Solving Integer Programs and relatives: a mostly discrete tour through optimization
Talk abstract: In this presentation, we will provide a comprehensive overview of mathematical optimization, covering a wide range of topics that may appear distinct, but share many commonalities. We will begin with an exploration of mixed-integer non-linear optimization (MINLP) and mixed-integer linear optimization (MILP), highlighting their importance in solving complex optimization problems. Our tour will then take a brief interlude to examine linear programming (LP) and Exact solutions, before delving into algorithms, software, modeling, and parallel computing. Along the way, we will delve into the intricacies of Steiner tree problems and explore the fascinating field of Quadratic Unconstraint Binary Optimization (QUBO), including its potential relevance to quantum computing. Finally, we will return to MILP, providing insights into solver development and offering a brief history of the SCIP Optimization Suite. Overall, this presentation aims to provide a comprehensive and insightful journey through the vast landscape of mathematical optimization, touching on key concepts, techniques, and developments in this dynamic field.
Bio: Prof. Thorsten Koch is Professor for Software and Algorithms for Discrete Optimization at TU-Berlin and serves as the head of the Applied Algorithmic Intelligence Methods and the Digital Data and Information for Society, Science, and Culture departments at the Zuse Institute Berlin (ZIB). He has worked in several areas, especially the planning of infrastructure networks, chip verification, mathematics education and integer optimization. Until 2020 he has been for two decades the head of the Integer Programming Solver Development Group at ZIB, which developed the SCIP solver during this time. From 2008-2014 he was the coordinator of the FORNE project, an industry collaboration project regarding gas transportation involving five universities and two research institutes. The project received the 2016 EURO Excellence in Practice Award of the European OR Society. From 2013-2019 he was head of the GasLab and the SynLab within the Research Campus MODAL (Mathematical Optimization and Data Analysis Laboratory). The project “Optimized Execution of Dispatching conducted together with Germanys largest Gas Transmission System Operator” became finalist of the 2020 INFORMS Innovative Applications in Analytics Award. Currently, the work is focused on developing high-performance parallel methods for solving large-scale structured optimization problems. Such problems arise, for example, in data-driven, real-world analysis and planning of sustainable network infrastructures. This includes high-performance solvers for Steiner Tree Problems in Graphs (STPG) and Quadratic Unconstraint Binary Optimization (QUBO).
Time: 4:00 – 5:00 PM, May 8th, 2023(Beijing time)
Zoom Meeting ID: 931 1530 3853 (Passcode: 530383)
Video Link: Solving Integer Programs and relatives: a mostly discrete tour through optimization