A decision space variable topology feature visualization and clustering method for Dynamic Multi-objective Optimization
Min Chen, Yanyan Zhang · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2024
We propose a dynamic multi-objective optimization algorithm for visualizing and clustering topological features of decision space variables. It maps all the obtained historical optimal solutions to the two-dimensional space through t-SNE, clusters the points, and selects a corresponding proportion of individuals from different clusters to form a new population. Throughout the search process, the topological characteristics of the reduced historical optimal solutions can be visually displayed. We conducted a preliminary experimental analysis on the standard test function to verify the effectiveness of the proposed method.