[1M2-CC] Synergy among Machine Learning, Computational Optimization, and Human Awareness / General Session – A

[1M2-CC-01] Enhancing multi-objective chaotic evolution algorithm using an estimated convergence point

Fengkai Guo¹, Yan Pei¹ (1. The University of Aizu)

In this paper, we attempt to use a method of estimating a convergence point of the population to accelerate the search of the multi-objective chaotic evolution optimization. The movement vectors between generations have powerful information for inducing the search direction of the global optimum solution. We use these movement vectors that are omposed of the non-dominated Pareto solutions to estimate a convergence point in which is the first Pareto front solution to enhance the search of multi-objective chaotic evolution algorithm. The estimated point is constricted by the movement vectors, and we use the estimated point to replace the population’s dominated solution to achieve the objective of enhancing the multi-objective chaotic evolution algorithm. We use hypervolume, generational distance, and inverted generational distance to evaluate our proposal. The result indicates that using an estimated point can accelerate the search of the multi-objective chaotic evolution algorithm.