A Large Scale Evolutionary Algorithm Based on Determinantal Point Processes for Large Scale Multi-Objective Optimization Problems

Global optimization challenges are frequent in scientific and engineering areas where loads of evolutionary computation methods i.e., differential evolution (DE) particle-swarm (PSO) employed to handle these problems. However, the performance algorithms declines due expansion problem dimension. The obstructed congregate with Pareto front rapidly while using large-scale algorithm. This work intends a multi-objective scheme aided by determinantal point process (LSMOEA-DPPs) this problem. proposed DPP model introduces mechanism consisting kernel matrix probability achieve convergence population variety high dimensional relationship balance keep diverse. We have also elitist non-dominated sorting for environmental selection. Moreover, projected algorithm demonstrates distinguishes four cutting-edge algorithms, each two three objectives, respectively, up 2500 decision variables. experimental results show that LSMOEA-DPPs outperform large margin.