韩耀宗教授受邀进行讲座

发布时间: 2024-01-05 14:35:00

题目：AI-Integration based Meshless Computational Methods for Solving Financial Option Pricing and Medical Imaging Problems

报告人：韩耀宗

时间：2023-01-04 15:00-16:00

地点：腾讯会议：656-4902-4546

报告人简介：

韩耀宗教授现任中国科学技术大学心理学系兼职教授，香港城市大学数学系客席教授，和香港优才计划元智智能研究所执行主任。已经发表SCI论文160余篇，被抄引超过8000篇次。John Ioannidis教授团队列为世界上被引用次数最多的2%科学家。同时，韩耀宗教授也是香港特别行政区政府教育局局内行政委员高级顾问委员会成员，曾获美国数学教育的知名专家，在教育研究领域发明使用修正的多元可能理论识别数学学习能力，以及为资优学生开设有效的STEM数学教育课程。

讲座摘要：

In this talk, an overview on global, local, quasi, residue, integration, and AI-based meshless computational methods via the use of radial basis function (RBF) and piece-wise polynomials will be presented. The main idea of the integration-based meshless computational method, Finite Integration Method (FIM), is to transform the original partial differential equation into an equivalent integral equation whose approximation can be sought by standard numerical integration techniques. Due to the advantages of unconditional stability of numerical integration and higher convergence rate of polynomial interpolation, the FIM method can solve boundary value problem under irregular domain with various kinds of stiffness. Recently, the FIM was combined with AI techniques to solve multi-dimensional problems such as 2D financial option pricing and 3D medical imaging.

主办单位：华南师范大学数学科学学院，机器学习与最优化计算实验室

协办单位：广东省青年科学家协会，广东省计算数学学会