Use of Traditional Optimization Ideas in Evolutionary Algorithms
04 April 2023 (Tuesday)
16:00 – 17:00 PM (Beijing Time, GMT+8)
Online via Tencent Meeting
Meeting ID: 672-707-922
Organized by the IEEE Computational Intelligence Society Shenzhen Chapter.
Activity Aim & Talk
In this virtual activity, Professor Qingfu Zhang will deliver a lecture on evolutionary algorithms (EAs). EAs have been widely used in many application field. However, EA Theory is quite poor and there is no theoretically solid principles for designing EAs. In this talk, I will introduce some of our attempts to understand and design EAs from the point of view of traditional math programming and statistics. Our philosophy is that evolutionary computation can treated as a branch of optimization, and thus commonly-used traditional optimization techniques can be used or modified for designing EAs. My talk includes four parts (1) EAs vs traditional optimization, (2) orthogonal crossover, (3) function smoothing, and (4) algorithm portfolio.
Meet the Speaker
Professor Qingfu Zhang, is a Chair Professor of Computational Intelligence with the Department of Computer Science, City University of Hong Kong. His is an IEEE fellow. His main research interests include evolutionary computation, optimization, metaheuristic, machine learning and their applications. He leads the AI and Meaheutistics Group and is supervising or co-supervising 17 PhD students and 4 postdoc research fellows in CityU. His MOEA/D algorithms have been widely researched and applied in industry.