Influence of initialization on the performance of metaheuristic optimizers

Self Cite

As an essential step of metaheuristic optimizers, initialization seriously affects the convergence speed and solution accuracy. The main motivation of the state-of-the-art initialization method is to generate a small initial population to cover the search space as much as possible uniformly. However, these approaches have suffered from the curse of dimensionality, high computational cost, and sensitivity to parameters, which ultimately reduce the algorithm's convergence speed. In this paper, a new initialization technique named diagonal linear uniform initialization (DLU) is proposed, which follows a novel search view, i.e., adopting the diagonal subspace sampling instead of the whole space. By considering the algorithm's update mechanism, the improved sampling method dramatically improves the convergence speed and solution accuracy of metaheuristic algorithms. Compared with the other eight widely used initialization strategies, the differential evolution (DE) algorithm with DLU obtains the best performance in search accuracy and convergence speed. In the extension experiments, results show that the DLU is still effective for three swarm-based algorithms: particle swarm optimization (PSO), cuckoo search (CS), and artificial bee colony (ABC). Especially for the multi-objective problem, the DLU still demonstrates its powerful performance compared with other strategies.

Section: A Comparison Of Different Initialization Methodsmentioning

confidence: 99%

Section: A Comparison Of Different Initialization Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Improved Initialization Method for Metaheuristic Algorithms: A Novel Search Space View

Bai

Gao

2021

Self Cite

“…The authors in [56] carried out a systematic comparison of the effect of 22 different probability distribution initialization methods on the convergence and accuracy of five optimization algorithms. Their results showed that the population size and maximum number of iterations affect most algorithms differently.…”

Section: Related Workmentioning

confidence: 99%

Influence of Initializing Krill Herd Algorithm With Low-Discrepancy Sequences

Agushaka

Ezugwu

2020

The krill herd (KH) algorithm is a global metaheuristic algorithm that was initially proposed for solving continuous optimization problems. The KH algorithm, since inception, has generated considerable real-world application interests in the research community. The standard algorithm solution implementation steps follow the initialization mechanism, which relies mainly on educated guesses or random initialization solution generation. Therefore, to improve the performance of the KH algorithm, the current study is set to investigate the influence of initializing the KH algorithm with three low-discrepancy sequences, such as the Faure sequence, Sobol sequence, and Van der Corput sequence. These low-discrepancy sequences are known to be more uniformly distributed across the problem search space than the commonly used random number initialization method. The study also evaluates the influence of population size on the performance of the proposed variants of the improved KH algorithms. The experimental results show significant improvements for the enhanced KH algorithms in terms of performance and the quality of solutions obtained; particularly on standard benchmarked high-dimension test problem instances, where the enhanced KH variants outperformed the existing basic KH algorithm for all the test functions evaluated. Similarly, the results for low dimension test cases showed less sensitivity to the initialization schemes, as the performance of our proposed improved scheme was comparable to that of the basic KH algorithm. However, in most cases, as the problem dimension was scaled up, the enhanced KH outperformed the basic KH. Evaluation results based on the population size of the algorithm, revealed that when the number of Krill is set at 25, the Sobol based KH initialization scheme performed better than did the other methods. Although, the Van der Corput and Faure based KH initialization schemes showed similar sensitivity when the dimension was set at 20. As we varied the population size of Krill, it was observed that the performance of the Sobol based KH initialization scheme deteriorated, whereas the other two methods showed superior performance. Overall, the findings from this study revealed that there are significant improvements in the performance of KH algorithm when initialized with low-discrepancy sequences.

“…Physics-inspired algorithms consist of simulated annealing [13], gravitational search algorithm [14], and quantum computing [15], while sociology-inspired ones usually denote imperialist competitive algorithm [16], brain storm optimization [17], culture algorithm [18], memetic algorithms [19], and so on. More importantly, these MHAs have been widely applied on various practical problems, from engineering [20], [21] to bio-informatics [22], [23], and achieved great successes in comparison with traditional mathematical analysis methods as they can obtain an acceptable solution with reasonable computational burden [24]- [28].…”

Section: Introductionmentioning

confidence: 99%

Comparative Study on Single and Multiple Chaotic Maps Incorporated Grey Wolf Optimization Algorithms

Yang

et al. 2021

As a meta-heuristic algorithm that simulates the intelligence of gray wolves, grey wolf optimizer (GWO) has a wide range of applications in practical problems. As a kind of local search, chaotic local search (CLS) has a strong ability to get rid of the local optimum due to its integration of chaotic maps. To enhance GWO, CLS is always incorporated into GWO to increase its population diversity and accelerate algorithm's convergence. However, it is still unclear that how may chaotic maps should be used in CLS and how to embed them into GWO. To address these challenging issues, this paper studies both single and multiple chaotic maps incorporated GWOs. Extensive comparative experiments are conducted based on IEEE Congress on Evolutionary Computation (CEC) benchmark test suit. The results show that CLS incorporated GWOs generally perform better than the original GWO, suggesting the effectiveness of such hybridization. Moreover, a remarkable finding of this work is that the piecewise linear chaotic map (PWLCM) and Gaussian map have the most potential to improve the search performance of GWO. Additionally, CLS incorporated GWOs also perform significantly better than some other state-of-the-art meta-heuristic algorithms. This study not only gives more insights into the mechanism of how CLS makes influence on GWO, but also finds that the most suitable choice of chaotic map for it.