学术报告201612期

报告题目：MANY-OBJECTIVE  EVOLUTIONARY ALGORITHMS FOR OPTIMIZATION

报 告 人：Gary  G. Yen, Regents Professor, FIEEE, FIET 

Oklahoma  State University  School of Electrical and Computer  Engineering

报告时间：2016年5月11日上午10:  00

报告地点：南一楼中311室

邀  请  方：“多谱信息处理技术”国家级重点实验室

Abstract:

Evolutionary computation is the study of biologically  motivated computational paradigms which exert novel ideas and inspiration from  natural evolution and adaptation. The  applications of population-based heuristics in solving multiobjective  optimization problems have been receiving a growing attention. To search for a  family of Pareto optimal solutions based on nature-inspiring problem solving  paradigms, Evolutionary Multiobjective Optimization Algorithms have been  successfully exploited to solve optimization problems in which the fitness  measures and even constraints are uncertain and changed over  time.

When encounter optimization problems  with many objectives, nearly all designs performs poorly because of loss of  selection pressure in fitness evaluation solely based upon Pareto optimality  principle.  This talk will survey recently published literature along this line of research-  evolutionary algorithm for many-objective optimization and its real-world  applications. Based on performance metrics ensemble, we will provide a comprehensive  measure among all competitors and more importantly reveal insight pertaining to  specific problem characteristics that the underlying evolutionary algorithm  could perform the best. The experimental results confirm the finding from the No  Free Lunch theorem: any algorithm’s elevated performance over one class of  problems is exactly paid for in loss over another class.

Biography:

Gary  G. Yen received the Ph.D. degree in electrical and computer engineering from the  University of Notre Dame in 1992. He is currently a Regents Professor in the  School of Electrical and Computer Engineering, Oklahoma State University. His  research interest includes intelligent control, computational intelligence,  evolutionary multiobjective optimization, conditional health monitoring, signal  processing and their industrial/defense applications.

         Gary was an  associate editor of the IEEE Transactions on Neural Networks and IEEE  Control Systems Magazine during 1994-1999, and of the IEEE Transactions  on Control Systems Technology, IEEE Transactions on Systems, Man and  Cyberneticsand IFAC Journal  on Automatica and Mechatronics during 2000-2010. He is currently  serving as an associate editor for the IEEE Transactions on Evolutionary  Computation and IEEE Transactions on Cybernetics. Gary served as Vice  President for the Technical Activities, IEEE Computational Intelligence Society  in 2004-2005 and is the founding editor-in-chief of the IEEE Computational  Intelligence Magazine, 2006-2009. He was the President of the IEEE  Computational Intelligence Society in 2010-2011 and is elected as a  Distinguished Lecturer for the term 2012-2014. He received Regents Distinguished  Research Award from OSU in 2009, 2011 Andrew P Sage Best Transactions Paper  award from IEEE Systems, Man and Cybernetics Society, 2013 Meritorious Service  award from IEEE Computational Intelligence Society and 2014 Lockheed Martin  Aeronautics Excellence Teaching award. He is a Fellow of IEEE and  IET.