Academic Report Notice of Qianlong Dang (North West Agriculture and Forestry University): Graph Neural Network Assisted Evolutionary Algorithm for Solving Multimodal Multi-objective Optimization

发布者:王健发布时间:2025-04-10浏览次数:60

Speaker:Associate Prof. Qianlong Dang

Affiliation: College of Science, Northwest Agriculture and Forestry University, China

Title: Graph Neural Network Assisted Evolutionary Algorithm for Multimodal Multi-objective Optimization

Time: April 12, 2025 (Saturday) 16:00

Location: 290, Arts and Sciences Building

Abstract: In a multimodal multi-objective optimization problem (MMOP), multiple solutions in the decision space have similar objective values, and these solutions with good diversity provide rich choices for decision makers. Traditional evolutionary algorithms have developed various diversity maintenance mechanisms to ensure an adequate search of the decision space. However, these mechanisms do not fully utilize the hidden Pareto-optimal solution set (PS) in historical data, and thus they are less adaptive and context-aware. Inspired by data-driven evolutionary algorithms, the use of graph neural network models to learn the knowledge hidden in historical data and assist evolutionary algorithms to explore PSs with different distributions is a very promising research. Therefore, this presentation will introduce the research on neural network-assisted evolutionary algorithms for solving MMOPs, including the shortcomings of traditional evolutionary algorithms for solving MMOPs, the design of evolutionary algorithms based on graph neural networks, and the assistance of graph learning to the evolutionary process.

Personal Introduction:Qianlong Dang, Ph.D., M.S. Supervisor, Associate Professor, College of Science, Northwest A&F University. He is the deputy director of the Joint Laboratory of Low Altitude Aircraft Algorithms and Simulation of Northwest Agriculture and Forestry University, and has conducted research on evolutionary algorithms, federated learning, large language models and their engineering applications. He focuses on evolutionary algorithms, federated learning, large language models and their engineering applications, and has achieved a number of innovative research results. He has been published in IEEE Transactions on Evolutionary Computation, IEEE Computational Intelligence Magazine, IEEE Transactions on Geoscience and Remote Sensing, IEEE Internet of Science and Technology, and IEEE Internet of Science and Technology. IEEE Transactions on Geoscience and Remote Sensing, IEEE Internet of Things Journal, IEEE Transactions on Artificial Intelligence, IEEE Transactions on Big Data, IEEE Transactions on Emerging Topics in Computational Intelligence. Topics in Computational Intelligence, Evolutionary Computationand other international journals, published more than 30 academic papers, including 25 articles as the first author and corresponding author, with a total impact factor of more than 200, of which 11 articles in IEEE journals, the Chinese Academy of Sciences, the first region paper 15 articles. He has applied for 12 invention patents, and the applicant has won the second prize of excellent papers for young teachers of Shaanxi Mathematical Society in 2023 and the third prize of excellent papers for young teachers of Shaanxi Society of Industrial and Applied Mathematics. In addition, he presided over the top grant of China Postdoctoral Science Foundation and the youth project of Shaanxi Provincial Natural Science Foundation.