Performance Analysis and Comparison of Four Conventional Multi-objective Optimization Algorithms

With the development of Artificial Intelligence and Big Data, more and more heuristic multi-objective algorithms are applied to the training process of data sets. In this paper, four heuristic multi-objective optimizations which were widely used in the data set training in recent years, are selected for the performance analysis and comparison. Through the comparison and analysis of the performance index for these algorithms on the benchmark problem, the advantages and disadvantages of these strategies in ensuring the convergence of the algorithm and maintaining the diversity of the solution sets are systematically expounded. The simulation results show that these algorithms have their own advantages and disadvantages in solving different specific problems, and the setting of the parameters and the initialization of the solution sets will have a great impact on the performance of the algorithm. Moreover, the different methods have the different abilities in maintaining the convergence and diversity of the solution sets. Although the complex optimization method has a better solution effect, the calculation time cost is higher. In practical application, it is necessary to select the appropriate algorithm flexibly according to the actual problems and conditions.