Enhanced Gaussian Bare-Bone Imperialist Competition Algorithm Based on Doubling Sampling and Quasi-oppositional Learning for Global Optimization

Dongge Lei,Lulu Cai,Fei Wu

DOI: https://doi.org/10.1007/s44196-024-00503-x

IF: 2.259

2024-05-23

International Journal of Computational Intelligence Systems

Abstract:Gaussian bare-bone imperialist competitive algorithm (GBB-ICA) is an effective variant of imperialist competitive algorithm (ICA), which updates the position of colonies by sampling a Gaussian distribution. However, the mean and standard deviation adopted by GBB-ICA is calculated only using the positions of imperialist and the colony itself, making the searching tends to trap into local optimum. To overcome this drawback, a new double Gaussian sampling strategy is proposed in this paper. An extra Gaussian sampling point, whose mean and standard is calculated using the positions of the second best colony and the current colony itself, is introduced into GBB-ICA. To further speed up the convergence and explore informative region, the quasi-oppositional learning technique is incorporated into GBB-ICA to produce more potential candidates in the assimilation step as well as generating a higher quality initial population. The proposed algorithm is called quasi-oppositional learning-based double Gaussian sampling bare-bone imperialist competitive algorithm (QOLBDGSBB-ICA) and is tested on 20 benchmark functions and four engineering design problems. Experimental results show that the proposed algorithm outperforms over other referenced ICA variants on 19 benchmark functions, which well validates the effectiveness of the proposed algorithm.

computer science, artificial intelligence, interdisciplinary applications

What problem does this paper attempt to address?

This paper proposes an improved Gaussian-based bare-bones imperialist competitive algorithm (GBB-ICA), referred to as Quadrant Opposite Learning and Dual Gaussian Sampling based Bare-bones Imperialist Competitive Algorithm (QOLBDGSBB-ICA), for global optimization problems. The original GBB-ICA updates the positions of colonies using Gaussian distribution, but it is prone to falling into local optima. To address this issue, the paper introduces a new dual Gaussian sampling strategy by adding an additional Gaussian sample point, with its mean and standard deviation based on the positions of the second-best colony and the current colony. Additionally, quasi-opposite learning is introduced to accelerate convergence speed and explore more informative search areas, generating high-quality initial populations and more potential candidate solutions.

In the experimental section, QOLBDGSBB-ICA is tested on 20 benchmark functions and four engineering design problems, and the results show that the algorithm outperforms other ICA variants in 19 benchmark functions, demonstrating its effectiveness and optimization ability. Through the use of dual Gaussian sampling and quasi-opposite learning, the new algorithm improves search capability, avoids premature convergence, and better explores regions that may contain the global optimum solution.