报告人：Jacek Mandziuk教授 

工作单位：华沙理工大学（波兰）

报告题目：安全问题中的进化优化

报告时间：2022年11月10日（周四）15:00-16:00

报告链接：Teams Link

https://teams.microsoft.com/l/meetup-join/19%3aB4gmRcUATAMA2iJqi-xXvtfPFfTbxVJPxSW_pcAPBao1%40thread.tacv2/1638719716825?context=%7b%22Tid%22%3a%2222804ebb-30d5-47df-942f-f3a3722f0225%22%2c%22Oid%22%3a%2216a60c03-ad7a-4b85-a403-8ebd947e010c%22%7d 

内容摘要：

国土安全威胁的全球化，如国际恐怖主义、武器或毒品走私，以及大规模盗窃，已成为21世纪安全部队面临的主要挑战之一。近年来已经提出和发展了新的有科学基础的打击有组织犯罪和恐怖主义的方法。安全博弈（SG）是快速发展的研究领域之一，它以安全部队和有组织的攻击者之间的博弈形式来模拟战术安全问题。本次报告将首先介绍元启发式方法，即在具有不完全信息的一般和序列安全博弈中逼近最优防御者的策略。该方法（Evolutionary Approach to Security Games- EASG）采用了具有专门设计的染色体和遗传算子的进化算法。实验评估表明，EASG在时间和内存上的尺度优于最先进的混合整数-线性程序方法，同时在绝大多数情况下提供最优解或次优解。此外，EASG可以在任意时刻终止后仍能提供一个相当好的解，这使得它非常适用于受时间限制的SG应用。同时也将介绍对基线EASG公式的一些加强操作，包括添加模因操作（局部优化程序）、协同进化设置和神经进化方法。随后，将讨论其在各种安全场景下的应用，如仓库巡逻、渡船保护、偷猎预防和网络安全等。

个人简介：

Jacek Mandziuk教授，博士，理学博士，分别于1989年、1993年以及2000年获得波兰华沙理工大学（WUT）应用数学（荣誉）硕士学位、博士学位以及波兰科学院计算机科学博士学位。2011年，Jacek Mandziuk教授被授予正级教授的称号，目前担任数学和信息科学学院的正教授，人工智能和计算方法系主任以及计算机科学博士课程主任。Jacek Mandziuk教授现已发表3部专著和180+篇研究论文，曾担任2021年在波兰克拉科夫举行的IEEE进化计算大会的总联合主席以及2013-2022年IEEE SSCI人类计算智能年度研讨会主席。Jacek Mandziuk教授是富布赖特高级研究奖（加州大学伯克利分校和美国ICSI伯克利分校）和罗伯特·舒曼基金会奖学金（CNRS，贝桑松，法国）的获得者，曾担任新加坡南阳科技大学（2015-2017年）、新南威尔士大学（澳大利亚，2013年）、延世大学（韩国，2011年）和阿尔伯塔大学（加拿大，2011年）的客座教授。Jacek Mandziuk教授目前是IEEE的高级会员和IEEE ETTC人类智能任务组的创始主席，曾担任《IEEE Transactions on Neural Networks and Learning Systems》、《 IEEE Transactions on

Computational Intelligence and AI in games》以及《ACM Computing Surveys》的副编辑。Jacek Mandziuk教授目前的研究方向包括计算智能和人工智能方法在游戏中的应用，动态和双层优化问题，问题求解过程中的人机合作，以及开发通用的人类学习和解决问题的方法。想要了解更多信息，请访问http://www.mini.pw.edu.pl/~mandziuk 

【编辑：王健】

英文版：

Academic Report Notice of Jacek Mańdziuk：Evolutionary Methods in Security Issues

Speaker: Professor  Jacek Mańdziuk

Title: Evolutionary Methods in Security Issues

Time: 15:00-16:00, November 10th, 2022 (Thursday)

Website: Teams Link

https://teams.microsoft.com/l/meetup-join/19%3aB4gmRcUATAMA2iJqi-xXvtfPFfTbxVJPxSW_pcAPBao1%40thread.tacv2/1638719716825?context=%7b%22Tid%22%3a%2222804ebb-30d5-47df-942f-f3a3722f0225%22%2c%22Oid%22%3a%2216a60c03-ad7a-4b85-a403-8ebd947e010c%22%7d 

Abstract: 

    Globalization of threats to homeland security, such as international terrorism, smuggling of weapons or drugs, or large-scale thefts became one of the main challenges for security forces in the 21st century. In effect, new scientifically grounded methods for fighting organized crime and terrorism have been proposed and developed in recent years. One of the rapidly developing research areas are Security Games (SG), which model tactical security issues in the form of games played between security forces and  organized attackers.In this talk I will introduce our recent metaheuristic approach to approximation of the optimal defender’s  strategy in general-sum sequential Security Games with imperfect information. The method (Evolutionary Approach to Security Games - EASG) employs Evolutionary Algorithms with specially designed  chromosomes and genetic operators. Experimental evaluation indicates that EASG scales in time and  memory better than state-of-the-art Mixed Integer-Linear Program methods while providing optimal or close-to-optimal solutions in the vast majority of the cases. Furthermore, EASG can be terminated in any moment and still provide a reasonably good solution, which makes it particularly well suited for time-critical SG applications.Several enhancements to the baseline EASG formulation will also be presented, including the addition of memetic operations (local optimization procedures), co-evolutionary setup, and neuro-evolutionary approach. Applications in various security scenarios such as warehouse patrolling, ferries protection, poaching prevention and cybersecurity will be discussed.

Personal Introduction:

     Prof. Jacek Mańdziuk, Ph.D., D.Sc., received M.Sc. (Honors) and Ph.D. in Applied Mathematics from the Warsaw University of Technology (WUT), Poland in 1989 and 1993, resp., and D.Sc. degree in Computer Science from the Polish Academy of Sciences in 2000. In 2011 he was awarded the title of Professor Titular. He is a full professor at the Faculty of Mathematics and Information Science, WUT, Head of Division of Artificial Intelligence and Computational Methods, and Head of Doctoral Program in Computer Science at this faculty. He is the author of 3 books and 180+ research papers. He was General Co-Chair of the 2021 IEEE Congress on Evolutionary Computation, Krakow, Poland, and Chair of the annual IEEE SSCI Symposium on Computational Intelligence for Human-like Intelligence 2013-2022. Prof. Mańdziuk was a recipient of the Fulbright Senior Research Award (UC Berkeley and ICSI Berkeley, USA) and the Robert Schuman Foundation Fellowship (CNRS, Besancon, France). He was a visiting professor at Nanyang Technological University in Singapore (2015-2017), University of New South Wales (Australia, 2013), Yonsei University (South Korea, 2011) and University of Alberta (Canada, 2011). He is Senior Member of IEEE and Founding Chair of the IEEE ETTC Task Force on Toward Human-like Intelligence. He serves/served as an Associate Editor for the IEEE Transactions on Neural Networks and Learning Systems, the IEEE Transactions on Computational Intelligence and AI in games, and the ACM Computing Surveys. His research interests include application of Computational Intelligence and Artificial Intelligence methods to games, dynamic and bilevel optimization problems, and human-machine cooperation in problem solving. He is also interested in the development of general-purpose human-like learning and problem-solving methods. For more information please visit http://www.mini.pw.edu.pl/~mandziuk 

[Editor:Jian Wang]