彩神x快三

学术报告
我的位置在: 首页 > 学术报告 > 正文
A New Optimization Paradigm to Solve High-dimensional Expensive Problems
浏览次数:日期:2021-01-12编辑:信科院 科研办

报告人:Mengchu Zhou美国国家发明家科学院院士、IEEE Fellow、IFAC Fellow、AAAS Fellow、CAA Fellow,美国新泽西理工学院杰出教授

报告时间:2021年1月13日 (星期三) 上午10:00 - 11:30

报告地点:Zoom在线会议

https://us02web.zoom.us/j/2810019605?pwd=S09LNnl5dHdXajZBbEJJOVd4TVlmUT09

Meeting ID: 281 001 9605

Passcode: HNU2020 

欢迎广大师生参加!烦请大家提前安装Zoom会议软件,软件下载地址:

 

报告摘要:High-dimensional computationally expensive problems (HEPs) in which a single fitness evaluation consumes hours or even days have attracted increasing attention from both academia and industry. Exponentially expanding search space and complex landscape make HEPs extremely challenging to be solved by traditional algorithms with limited computational resources. Therefore, an Autoencoder-embedded Evolutionary Optimization (AEO) framework is invented to deal with them. To be specific, high-dimensional search space can be compressed to informative low-dimensional space by using an autoencoder as a dimension reduction tool. The search operation conducted in this low-dimensional space facilitates the population in convergence towards the optima. To balance the exploration and exploitation ability during optimization, two sub-populations are adopted to coevolve in a distributed fashion, wherein one is assisted by an autoencoder and the other undergoes a regular evolutionary process. Dynamic information exchange is conducted between them after each cycle to promote sub-population diversity. Moreover, surrogate models can be incorporated into AEO (SAEO) to further boost its performance by reducing unnecessary fitness evaluation. Both AEO and SAEO are validated by testing benchmark functions with dimensions varying from 30 to 200. Compared with the state-of-the-art algorithms for HEPs, AEO shows extraordinarily high efficiency for these challenging problems while SAEO can greatly improve the performance of AEO in most cases, thus opening new directions for various evolutionary algorithms under AEO to tackle HEPs and greatly advancing the field of high-dimensional computationally expensive optimization.



报告人简介:周孟初教授是美国新泽西理工学院的杰出教授,1995年获终身教职。现为美国国家发明家科学院院士、IEEE Fellow、IFAC Fellow(国际自动控制联合会会士)、AAAS Fellow(美国科学促进会会士)、CAA Fellow(中国自动化学会会士)。自1990年起在新泽西理工大学电气与计算机工程系任教,从事Petri网理论与工程应用,智能自动化、工业4.0、物联网、人工智能、大数据分析、云服务计算、边缘计算等方面的研究。周教授总共发表了900余篇期刊、会议论文,其中包括12本专_x0008_著,450余篇IEEE Trans. 文,28项国际专_x0008_利。在十年的时间里得到了美国国家科学基金(NSF)、国防部、美国国家标准技术院、国家航空航天署及新泽州科委等政府部门以及工程基金会及十多家公司的一千二百多万美元的研究资助,主持并参与了五十多个研究课题。周博士获得了许多嘉奖,主要有1994年美国制造工程师协会颁发的“计算机集成制造系统大学领先奖”; 2000年德国洪堡基金会的美国资深科学家洪堡研究奖; 2010年IEEE Systems, Man and Cybernetics学会年会的Franklin V. Taylor 最佳论文奖; 2015IEEE Systems, Man and Cybernetics学会的Norbert Wiener2019年新泽西理工学院卓越研究勋章; 2020年新泽西研究与开发委员会爱迪生发明奖。周博士高被引作者并于 2012 列在工程领域的高被引作者第一名


邀请人:李肯立


联系人:陈建国