题目一：Efficient synchronous retrieval of OAM modes and AT strength using multi-task neural networks

内容简介：When transmitted through the atmospheric channel, OAM beams are influenced by the random fluctuations in the refractive index caused by atmospheric turbulence, resulting in phase distortion and intensity dispersion of the beams, leading to severe signal interference. Due to the high randomness of atmospheric turbulence, it is essential for OAM mode recognition methods to have good stability to ensure communication quality. We establish an equivalence relationship between the continuous dynamics system and the network unit RUEM, ensuring its stability through theoretical derivation and numerical experiments. We propose a multitask neural network model, OATNN, embedded with RUEM to achieve efficient simultaneous recognition of turbulence intensity in atmospheric turbulence environments and OAM modes in free-space optical communication systems.

报告人：尹伟石

报告人简介：长春理工大学数学与统计学院副教授，硕士研究生导师。主要研究兴趣是数学物理反问题、机器学习算法的设计与理论分析和微分方程数值解等。在JCP，JCAM，CICP，IPI等期刊发表论文20余篇，主持并参与国家自然科学基金、吉林省科技厅基金和吉林省教育厅基金6项。目前担任中国仿真学会不确定系统分析与仿真专委会委员、Math Review评论员以及IPIA会员。

题目二：An online interactive physical information adversarial network for solving mean-field games

内容简介：In this talk, We propose an online interactive physical information adversarial network (IPIAN) to solve mean-field games (MFGs) from the perspective of physical information interaction. We consider the variational dual structure of MFGs, treat the interactions between agents as physical information interactions, simulate the evolution of individual strategy choices and the overall distribution of agents, and then use adversarial networks to solve MFGs.Based on the generation of an adversarial framework, we use two online physical information networks to solve the value function and the density function and train the networks to approximate the solution of MFGs by adversarial means. A self-attention mechanism is introduced to focus on strategic physical information to improve the expressiveness and accuracy of IPIAN. Numerical experiments demonstrate the effectiveness of IPIAN in solving high-dimensional mean-field game models by performing obstacle avoidance experiments on quadrotors in different contexts.

报告人：孟品超

报告人简介：长春理工大学数学与统计学院教授，硕士研究生导师。主要研究兴趣是数学物理反问题、机器学习算法的设计与理论分析等。在JCAM，OE，CICP，IPI等期刊发表论文20余篇，主持并参与国家自然科学基金、军委科技委领域基金、吉林省科技厅基金、吉林省教育厅基金8项。目前担任中国仿真学会不确定系统分析与仿真专委会委员、IPIA会员、吉林省运筹学会理事。

时  间：2024年8月9日（周五）下午16：00开始

地  点：南海楼330室

热烈欢迎广大师生参加!

信息科学技术学院

2024年8月8日