Speaker: Zhan Zhihui(Nankai University)
Title:Data and Knowledge Driven Evolutionary Computation
Time: 10:30, Friday, September 4, 2026
Location:254 Arts and Science Building
Abstract:
Evolutionary computation (EC) is a class of powerful artificial intelligence (AI) methods for optimization. It simulates evolutionary phenomena and swarm intelligent behavior in nature, and holds broad application prospects in knowledge creation and problem solving. EC algorithms follow Darwin’s principle of survival of the fittest to select superior solutions and reproduce new ones. However, they run into bottlenecks when dealing with expensive optimization problems — problems where fitness evaluation is extremely time-consuming and costly, or even the fitness function itself cannot be explicitly formulated. Complex optimization problems also pose challenges to EC algorithms: they are prone to falling into local optima, or take excessively long to converge to the region of promising solutions. Accordingly, data-driven EC (DDEC) and knowledge-driven EC (KDEC) have gradually become research frontiers, helping EC algorithms tackle these demanding optimization problems. This lecture will elaborate on the driving targets and implementation approaches of data/knowledge-driven evolutionary computation. In terms of driving targets, research mainly falls into two directions: one is to build surrogates for fitness evaluation to drive the selection operation; the other is to learn patterns from high-quality solutions to assist new solution generation, thus driving the evolutionary process. Specifically, for data-driven methods targeting the selection phase, the lecture will introduce the Boosting Data-Driven Evolutionary Algorithm (BDDEA) and the Hierarchical and Ensemble Surrogate-assisted Evolutionary Algorithm (HES-EA); for data-driven methods targeting the evolution phase, it will present Learning-aided Evolution for Optimization (LEO) and Knowledge Learning for Evolutionary Computation (KLEC). We expect that such new EC paradigms can provide fresh insights for solving modern ultra-complex optimization problems and advance the further development of evolutionary computation and artificial intelligence.
Personal Introduction:
Zhan Zhihui is a professor and doctoral supervisor at the College of Artificial Intelligence, Nankai University. He is an IEEE Fellow, recipient of the IEEE Computational Intelligence Society Outstanding Young Investigator Award (selected globally one per year), Yangtze River Scholar (Young Scholar) of the Ministry of Education of China, awardee of the National Science Fund for Distinguished Young Scholars, recipient of the Wu Wenjun Outstanding Young Artificial Intelligence Award, and a Clarivate Highly Cited Researcher. He ranks among the top 2% of scientists worldwide in the field of artificial intelligence (listed in both the Annual Scientific Impact and Lifetime Scientific Impact lists), and has been named a China Highly Cited Researcher for 12 consecutive years from 2014 to 2025. His main research areas cover artificial intelligence, evolutionary computation, swarm intelligence and their applications, and he serves as an associate editor of four top IEEE Transactions journals in the fields of evolutionary computation, artificial intelligence and control: IEEE Transactions on Evolutionary Computation, IEEE Transactions on Emerging Topics in Computational Intelligence, IEEE Transactions on Systems, Man, and Cybernetics: Systems, and IEEE Transactions on Artificial Intelligence.
[Editor:Zhang Yaoguang]