Speaker: Chen Changbo is a researcher, doctoral supervisor, member of the Chinese Society of Computer Mathematics, and director of the Mathematics and Intelligence Branch of the Operations Research Society of China.

Date: October 29, 2024

Time: 3:00-4:00 pm

Location: Tecent Meeting: 574 566 527

Sponsor: School of Mathematics, Shandong University

Abstract:

It is a daunting task to develop an efficient polynomial system solver leveraging modern hardware architectures, such as multi-core CPUs and GPUs. Automatic loop parallelization and auto-tuning provide a way to efficiently implement the basic kernels underlying the solers, such as matrix multiplication, polynomial multiplication, stencil computation, and so on. On the other hand, some key steps in such optimization, such as data dependence analysis, schedule computation, and code generation, require to efficiently solve polynomial (linear) systems. In this talk, I will explain the interactions between polynomial system solving and automatic loop optimization, and present our recent works on these interactions.

For more information, please visit:

https://www.view.sdu.edu.cn/info/1020/196257.htm