Wang Zhenzhong's doctoral defense focused on the research topic "Knowledge-Driven Evolutionary Algorithms for Multiobjective Optimization Problems." During his PhD journey, Wang Zhenzhong was devoted to studying knowledge-driven evolutionary algorithms for solving complex optimization problems such as large-scale optimization problems, dynamic optimization problems, and multitask optimization problems. His innovative achievements include:

1) Proposing a knowledge-driven offspring generation operator for large-scale optimization problems. In particular, this work leverages the manifold assumption to generate high-quality offspring solutions, improving the convergence and diversity of the population.

2) Proposing a knowledge-driven resource allocation strategy for dynamic optimization problems. Specifically, a Monte Carlo tree-assisted multi-population algorithm is devised to adaptively activate promising subpopulations and adopt proper evolutionary operators according to the changing environment, reducing computational resource waste.

3) Proposing a spatial-temporal knowledge transfer framework for dynamic-constrained optimization problems. Particularly, this work designs spatial and temporal knowledge transfer modules to maintain diversity, convergence, and feasibility. Additionally, to advance the test suite toward real-world cases, the work proposes 14 test problems with various properties.

4) Proposing a knowledge-driven solution selection method for evolutionary multitasking. This work leverages task-specific knowledge of source and target tasks to transfer solutions that can adapt well to the target task, thus promoting the positive transfer at large.

5) Proposing a knowledge-driven evolutionary solver for fairness debugging in graph data. Particularly, the work formulates fairness debugging as an expensive multiobjective optimization task. Then, a dedicated evolutionary solver is developed to produce compact, diagnostic, and actionable explanations for unfairness.

 

Dr. Wang Zhenzhong received his B.E. degree from Northeastern University, Shenyang, China, in 2017 and his M.E. degree from Xiamen University, Xiamen, China, in 2021. His research interests include AI for Science and evolutionary computation. He has published several papers in prestigious research journals and conferences, including TEVC, TNNLS, TCYB, AAAI, etc.

​

Dr. Wang Zhenzhong will join Xiamen University as an Assistant Professor, where he will continue his research in evolutionary optimization and AI for Science. We wish him a successful and enriching academic journey ahead!