Self-Adaptation Mechanism to Control the Diversity of the Population in Genetic Algorithm

TL;DR

This paper introduces a self-adaptive mechanism for controlling population diversity in genetic algorithms, eliminating the need for problem-specific parameter tuning and improving convergence to optimal solutions.

Contribution

It proposes a novel diversity control method based on mating preference competition that adapts automatically without explicit parameter setting.

Findings

Performance comparable to well-tuned traditional GAs

Effective in avoiding premature convergence

Applicable across multiple test problems

Abstract

One of the problems in applying Genetic Algorithm is that there is some situation where the evolutionary process converges too fast to a solution which causes it to be trapped in local optima. To overcome this problem, a proper diversity in the candidate solutions must be determined. Most existing diversity-maintenance mechanisms require a problem specific knowledge to setup parameters properly. This work proposes a method to control diversity of the population without explicit parameter setting. A self-adaptation mechanism is proposed based on the competition of preference characteristic in mating. It can adapt the population toward proper diversity for the problems. The experiments are carried out to measure the effectiveness of the proposed method based on nine well-known test problems. The performance of the adaptive method is comparable to traditional Genetic Algorithm with the best parameter setting.