A study on the use of non-parametric tests for analyzing the evolutionary algorithms’ behaviour: a case study on the CEC’2005 Special Session on Real Parameter Optimization

By Salvador Garcia, Daniel Molina, Manuel Lozano, Francisco Herrera. J. Heuristics, 2009

1,707 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

View Paper: https://www.semanticscholar.org/paper/3a8de0aa2b8b86cc0302d62257c23cc8c37d0880

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

Citations in the corpus

Alternative Restart Strategies for CMA-ES (2012, methodology): “Two versions of CMA-ES with restarts have been proposed to handle multi-modal functions: IPOP-CMA-ES [2] was ranked first on the continuous optimization benchmark at CEC 2005 [4, 3]; and BIPOP-CMA-ES [5] showed the best results together with IPOP-CMA-ES on the black-box optimization benchmark (BBOB)”
Lévy Flight Trajectory-Based Whale Optimization Algorithm for Global Optimization (2017, methodology): “Moreover, in this paper, we also conduct Wilcoxon’s nonparametric statistical test [31], [32] over these 30 runs to have a statistically meaningful conclusion.”
Dendritic Neuron Model With Effective Learning Algorithms for Classification, Approximation, and Prediction (2019, methodology): “Wilcoxon signed-ranks test [80] is used to detect significant”
Chaotic grasshopper optimization algorithm for global optimization (2019, result): “This means a statistical test is mandatory in order to verify that the proposed algorithm shows a noteworthy enhancement in comparison with other algorithms [47].”
A swarm optimization algorithm inspired in the behavior of the social-spider (2013, methodology): “of 400 iterations on average while the other three algorithms need much more iterations. A non -parametric statistical significance proof known as the Wilcoxon’s rank sum test for independent samples [43,44] has been conducted over the “average best-so-far” (AB) data of Table 2, with an 5% significa”
Diversity enhanced particle swarm optimization with neighborhood search (2013, methodology): “To compare the performance differences among DNSPSO and the other four PSO algorithms, we conduct a Wilcoxon signed-rank test [12,19].”
A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms (2011, methodology): “Instatisticalterms,weare losing the control on the Family-Wise Error Rate (FWER), defined as the probability of making one or more false discoveries among all the hypotheses when performing multiple pairwise tests.”
Advanced nonparametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining: Experimental analysis of power (2010, background): “However, some studies involving CI algorithms in experimental comparisons show that these conditions are not easy to meet [21] [23] [47].”
Harris hawks optimization: Algorithm and applications (2019, methodology): “The 218 third group problems (F24-F29) are selected from IEEE CEC 2005 competition [51] and covers 219 hybrid composite, rotated and shifted MM test cases.”
Let a biogeography-based optimizer train your Multi-Layer Perceptron (2014, background): “It is not enough to compare algorithms based on the mean and standard deviation values [15]; a statistical test is necessary to prove that a proposed new algorithm presents a significant improvement over other existing methods for a particular problem [38].”
A Fuzzy Association Rule-Based Classification Model for High-Dimensional Problems With Genetic Rule Selection and Lateral Tuning (2011, methodology): “ple comparison [40]–[43] of the performance of these classifiers.”
S-shaped versus V-shaped transfer functions for binary Particle Swarm Optimization (2013, background): “In other words, it is not enough to compare algorithms based on the mean and standard deviation values [32] and a statistical test is necessary to prove that a proposed new algorithm presents a significant improvement over other existing methods for a particular problem.”