On Selection of a Benchmark by Determining the Algorithms’ Qualities

Fister, Iztok; Brest, Janez; Iglesias, Andrés; Gálvez, Antonio; Deb, Suash

doi:10.1109/access.2021.3058285

Cited by 13 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various kinds of Swarm Intelligence and Evolutionary Algorithms have been compared multiple times in the literature, both theoretically and empirically, but we are not aware of papers in which a relative improvement of many novel algorithms over the classical ones would be addressed. In numerous guides a number of rules have been proposed that, at least in the opinion of their authors, should be followed when comparing various algorithms [39], [40], [41], [42]. In practice often it is considered that the safest choice is to follow the rules defined for some widely accepted benchmark sets, such as IEEE or BBOB ones [33], [34].…”

Section: Literature Reviewmentioning

confidence: 99%

How Much Do Swarm Intelligence and Evolutionary Algorithms Improve Over a Classical Heuristic From 1960?

2023

View full text Add to dashboard Cite

Hundreds of variants of Swarm Intelligence or Evolutionary Algorithms are proposed each year and numerous competitions and comparisons between algorithms may suggest rapid improvement in the field. However, such comparisons are often done between a limited number of methods and are based on averaged ranks of algorithms. This way they measure whether one method is on average ranked better than the others, without giving any information on how much improvement is in fact obtained. In this study we show a general comparison between 69 algorithms, starting from methods proposed in the 1960's up to variants developed in the early 2020's, on single-objective static numerical problems. Algorithms are compared on searching for a minimum of 30 different 50-dimensional mathematical functions, and on 22 real-world problems. We focus on the relative improvement achieved by various algorithms over a single-solution based method proposed in 1960 by Howard Rosenbrock. We find that the general improvement of Evolutionary Algorithms over Rosenbrock's algorithm is relatively limited. It is high for the artificial benchmarks, for which many Evolutionary Algorithms find solutions 10 times closer to the global optimum in terms of fitness than Rosenbrock's algorithm, but much lower for real-world problems. Improvement is also higher when performance averaged over many runs is compared, but lower when the best results from multiple runs are analyzed. In the last case, only the best Evolutionary Algorithms are able to find solutions of a ''typical'' real-world problem that are 2-3 times better in terms of fitness than those found by Rosenbrock's algorithm. The relative improvement of recently proposed algorithms is not much better than the improvement achieved by algorithms proposed over a decade ago.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

How Much Do Swarm Intelligence and Evolutionary Algorithms Improve Over a Classical Heuristic From 1960?

2023

View full text Add to dashboard Cite

show abstract

“…With the increasing number of nature-inspired algorithms, various benchmarking tests have been developed to examine their performance [46]. These include testing the algorithms on different types of functions [47], [48], and checking the number of objective function evaluations they use [49].…”

Section: B Nature-inspired Algorithmsmentioning

confidence: 99%

Philippine Eagle Optimization Algorithm

2022

View full text Add to dashboard Cite

We propose the Philippine Eagle Optimization Algorithm (PEOA), which is a meta-heuristic and population-based search algorithm inspired by the territorial hunting behavior of the Philippine Eagle. From an initial random population of eagles in a given search space, the best eagle is selected and undergoes a local food search using the interior point method as its means of exploitation. The population is then divided into three subpopulations, and each subpopulation is assigned an operator which aids in the exploration. Once the respective operators are applied, the new eagles with improved function values replace the older ones. The best eagle of the population is then updated and conducts a local food search again. These steps are done iteratively, and the food searched by the final best eagle is the optimal solution of the search space. PEOA is tested on 20 optimization test functions with different modality, separability, and dimension properties. The performance of PEOA is compared to 13 other optimization algorithms. To further validate the effectiveness of PEOA, it is also applied to image reconstruction in electrical impedance tomography and parameter identification in a neutral delay differential equation model. Numerical results show that PEOA can obtain accurate solutions to various functions and problems. PEOA proves to be the most computationally inexpensive algorithm relative to the others examined, while also helping promote the critically endangered Philippine Eagle.

show abstract

“…In this section, IKMA is tested and compared with other algorithms by using 23 benchmark functions (Tables 2 and 3) from the CEC2013 [34]. The fitness function values' average and standard deviation (STD) are used as evaluation metrics to compare the algorithms' merits.…”

Section: A Cec2013 Benchmark Functionsmentioning

confidence: 99%

Improved Adaptive Komodo Mlipir Algorithm

Liu

Zhang

2022

IEEE Access

View full text Add to dashboard Cite

In order to improve the global search performance of the Komodo Mlipir Algorithm, this paper proposed two adaptive Komodo Mlipir Algorithms with variable fixed parameters (IKMA-1; IKMA-2). Among them, IKMA-1 adaptively controls the parthenogenesis radius of female Komodo dragons to achieve more efficient conversion of global search and local search. Second, IKMA-2 introduces adaptive weighting factors to the ''mlipir'' movement formula of Komodo dragons to improve the local search performance. Both IKMA-1 and IKMA-2 were tested on 23 benchmark functions in CEC2013 and compared with the other seven optimization algorithms. The Wilcoxon rank-sum test and Friedman rank test were used to compare the performance of different algorithms. Furthermore, IKMA-1 and IKMA-2 are applied to two constrained engineering optimization problems to verify the engineering applicability of the improved algorithm. The results show that both IKMA-1 and IKMA-2 have better convergence accuracy than the initial KMA. In terms of the benchmark function simulation results, IKMA-1 improves the performance by 17.58% compared to KMA; IKMA-2 improves by 10.99%. Both IKMA-1 and IKMA-2 achieve better results than other algorithms for engineering optimization problems, and IKMA-2 outperforms IKMA-1.INDEX TERMS Komodo Mlipir Algorithm, variable fixed parameters, adaptive optimization, engineering design optimization problems. I. INTRODUCTIONEstablishing models to deal with practical problems is an essential means of current academic research, and how to solve models faster and better depends on the actual solution performance of various algorithms. An optimization algorithm is an application technology based on mathematics and used to solve various practical optimization problems. At present, optimization algorithms can be mainly divided into data processing algorithms, neural network algorithms, and swarm intelligence algorithms [1]. Applied optimization problems widely exist in many fields, such as signal processing, production scheduling, medical applications, image processing, and path planning.However, since these optimization problems often involve discrete, discontinuous, and uncertain factors, it is not realistic to rely on a single algorithm to solve all optimization problems in life [2]. Therefore, the innovation of newThe associate editor coordinating the review of this manuscript and approving it for publication was Muhammad Zakarya .

show abstract

On Selection of a Benchmark by Determining the Algorithms’ Qualities

Cited by 13 publications

References 28 publications

How Much Do Swarm Intelligence and Evolutionary Algorithms Improve Over a Classical Heuristic From 1960?

How Much Do Swarm Intelligence and Evolutionary Algorithms Improve Over a Classical Heuristic From 1960?

Philippine Eagle Optimization Algorithm

Improved Adaptive Komodo Mlipir Algorithm

Contact Info

Product

Resources

About