Eigen Crossover in Cooperative Model of Evolutionary Algorithms Applied to CEC 2022 Single Objective Numerical Optimisation

Bujok, Petr; Kolenovský, Patrik

doi:10.1109/cec55065.2022.9870433

Cited by 31 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This subsection aims to reveal the competitive edge of RPSO by comparing it with top algorithms from different years of the CEC competition, including the gaining-sharing knowledge-based algorithm with adaptive parameters hybrid with IMODE (APGSK-IMODE) [62], Eigen crossover in cooperative model of evolutionary algorithms (EA4eig) [63], Improved multi-operator differential evolution (IMODE) [64], Population's variance-based adaptive differential evolution (PVADE) [65], and the adaptive gaining-sharing knowledgebased algorithm (AGSK) [66]. EA4eig was the champion algorithm in CEC 2022, APGSK-IMODE ranked fourth in CEC 2021, IMODE was the champion in CEC 2020, AGSK was the runner-up in CEC 2020, and PVADE was a recognized algorithm in the competition.…”

Section: Comparison With the Top Algorithmmentioning

confidence: 99%

A Random Particle Swarm Optimization Based on Cosine Similarity for Global Optimization and Classification Problems

Liu,

Zeng,

et al. 2024

Biomimetics

View full text Add to dashboard Cite

In today’s fast-paced and ever-changing environment, the need for algorithms with enhanced global optimization capability has become increasingly crucial due to the emergence of a wide range of optimization problems. To tackle this issue, we present a new algorithm called Random Particle Swarm Optimization (RPSO) based on cosine similarity. RPSO is evaluated using both the IEEE Congress on Evolutionary Computation (CEC) 2022 test dataset and Convolutional Neural Network (CNN) classification experiments. The RPSO algorithm builds upon the traditional PSO algorithm by incorporating several key enhancements. Firstly, the parameter selection is adapted and a mechanism called Random Contrastive Interaction (RCI) is introduced. This mechanism fosters information exchange among particles, thereby improving the ability of the algorithm to explore the search space more effectively. Secondly, quadratic interpolation (QI) is incorporated to boost the local search efficiency of the algorithm. RPSO utilizes cosine similarity for the selection of both QI and RCI, dynamically updating population information to steer the algorithm towards optimal solutions. In the evaluation using the CEC 2022 test dataset, RPSO is compared with recent variations of Particle Swarm Optimization (PSO) and top algorithms in the CEC community. The results highlight the strong competitiveness and advantages of RPSO, validating its effectiveness in tackling global optimization tasks. Additionally, in the classification experiments with optimizing CNNs for medical images, RPSO demonstrated stability and accuracy comparable to other algorithms and variants. This further confirms the value and utility of RPSO in improving the performance of CNN classification tasks.

show abstract

Section: Comparison With the Top Algorithmmentioning

confidence: 99%

A Random Particle Swarm Optimization Based on Cosine Similarity for Global Optimization and Classification Problems

Liu,

Zeng,

et al. 2024

Biomimetics

View full text Add to dashboard Cite

show abstract

“…T0 is estimated by evaluating mathematical functions. T1 is calculated by conducting 200000 evaluations of the 18th function in CEC2017, 1st function in CEC2020 [93], and CEC2022 [94] for all dimensions. T2 is obtained by averaging 5 algorithms executed on CEC18 (in CEC2017) and CEC01 (CEC2020 and CEC2022) with 200000 evaluations.…”

Section: E Complexity Analysismentioning

confidence: 99%

Humboldt Squid Optimization Algorithm (HSOA): A Novel Nature-Inspired Technique for Solving Optimization Problems

Anaraki,

Farzin

2023

IEEE Access

View full text Add to dashboard Cite

This study presents a new natural-based algorithm called the Humboldt squid optimization algorithm (HSOA). HSOA is inspired by Humboldt squids hunting, moving, and mating behavior. The HSOA search procedure involves an attack on fish schools, a fish's escape, a successful attack, an attack of bigger squids on smaller ones, and mating, which is the inspiration for creating an algorithm to address existing issues. In HSOA, half of the best populations are Humboldt squid, and the rest are school fish. Individuals connect with each other and cooperate to achieve the optimal response. HSOA is versatile and applicable to mathematical and engineering problems. Solving eighty-four benchmark function problems (twenty-three classic functions, twenty-nine CEC-BC-2017 with 10, 30, 50, and 100 dimensions, ten CEC-C06 2019, ten CEC2020 with 5, 10, 15, and 20 dimensions, and twelve CEC2022 with 10 and 20 dimensions) and twenty-four engineering problems (six CEC2006 and eighteen CEC2011) shows that our proposed algorithm provides proper and acceptable answers to nine algorithms, including well-known (PSO, DE, WOA), recent (AVOA, RW_GWO, HHO and GBO), and state-of-the-art algorithms (LSHADE and EBOwithCMAR). Friedman's rank from HSOA for one hundred and eight problems was 16.45% and 7.45% lower than LSHADE and EBOwithCMAR. Thus, HSOA has the potential to solve various complex problems in the sciences and engineering fields.

show abstract

“…The winner of the CEC 2022 competition EA4eig [7] is adapted from the ensemble of four population-based metaheruistics -CoBiDE [17], IDEbd [18], CMA-ES [19], and jSO [20] -proposed in [21] by extending the application range of the Eigen approach from CoBiDE to all the constituent algorithms. NL-SHADE-LBC [8] ranks second in the competition.…”

Section: Algorithms Under the New Settingmentioning

confidence: 99%

“…In the CEC 2021 competition, APGSK-IMODE [4], NL-SHADE-RSP [5], and jDE-21 [6] are the winners of different tracks. In the CEC 2022 competition, the best three performers are EA4eig [7], NL-SHADE-LBC [8], and NL-SHADE-RSP-MID [9]. More details of the above algorithms can be found in the next section -related work.…”

Section: Introductionmentioning

confidence: 99%

Ensemble of Population-Based Metaheuristics With Exchange of Individuals Chosen Based on Fitness and Lifetime for Real Parameter Single Objective Optimization

Zhu

et al. 2023

Preprint

View full text Add to dashboard Cite

Real parameter single objective optimization has been studied for decades to find the best decision vector to minimize (or maximize) an objective multi-dimension function. In recent, a new setting is proposed and applied in this field based on the consideration that solving difficulty scales exponentially with the increase of dimensionality. Under the new setting, Differential Evolution (DE) still outstands in performance as before. Meanwhile, a new type of population-based metaheuristic - Gaining-Sharing Knowledge based algorithm (GSK) becomes a dark horse. Furthermore, ensemble of the above two types of algorithm - APGSK-IMODE - is proposed in literature. In this paper, we change the scheme for executing communication between the two constituent algorithms - APGSK and IMODE, and propose APGSK-IMODE-II. In our experiment, our algorithm is compared with state-of-the-art algorithms. Experimental results show that our APGSK-IMODE-II is very competitive.

show abstract

Eigen Crossover in Cooperative Model of Evolutionary Algorithms Applied to CEC 2022 Single Objective Numerical Optimisation

Cited by 31 publications

References 24 publications

A Random Particle Swarm Optimization Based on Cosine Similarity for Global Optimization and Classification Problems

A Random Particle Swarm Optimization Based on Cosine Similarity for Global Optimization and Classification Problems

Humboldt Squid Optimization Algorithm (HSOA): A Novel Nature-Inspired Technique for Solving Optimization Problems

Ensemble of Population-Based Metaheuristics With Exchange of Individuals Chosen Based on Fitness and Lifetime for Real Parameter Single Objective Optimization

Contact Info

Product

Resources

About