A failure remember-driven self-adaptive differential evolution with top-bottom strategy

Dual Elite Groups-Guided Differential Evolution for Global Numerical Optimization

“…In the above manner, individuals in DE in the initial stage are expectedly dispersed in diverse areas of the solution space, which they can then search in distinct directions [19,[22][23][24][25].…”

“…Researchers' attention has been primarily focused on developing a uniform mutation scheme for all individuals to evolve. The key to this usually lies in the selection of parent individuals participating in the mutation operation [16,19,[23][24][25]. At first, researchers focused on employing topologies to organize individuals and then selecting parents based on the adopted topology structures to direct the mutation of each individual [11,12,15].…”

“…After that, the devised mutation scheme is performed to mutate the individual (line 14), following which the binomial crossover operation generates the final offspring, as shown in line 15. Afterwards, the offspring is evaluated and the records in the relevant archives are updated along with the selection operation (lines [16][17][18][19][20][21][22][23][24][25][26]. The above operations are performed iteratively until the termination condition is satisfied, which is commonly the completion of the maximum number of fitness evaluations.…”

“…In general, the mutation operation is usually realized by shifting the base individual by the difference vectors between individuals in the population [19,[22][23][24][25]. In essence, the key to designing effective mutation strategies lies in the selection of individuals participating in the mutation, such that the mutation diversity of the population remains high and, at the same time, the mutated individuals are likely to approach optimal areas fast [11,12,15,16,20].…”

Differential Evolution and Its Applications in Image Processing Problems: A Comprehensive Review

“…Analysis of the results affirmed robustness and efficiency ADE-ALC Fu et al [ 64 ] An aging mechanism was utilized to select control parameters with a probability distribution Test results revealed that ADE-ALC, compared with other DE variants using the unimodal, multimodal, hybrid function, was either better or more competitive FDE Ochoa et al [ 65 ] Dynamically selected parameters using fuzzy logic to improve exploration and exploitation Performed better compared to algorithms of its type LSHADE-cnEpsin Awad et al [ 44 ] Introduced an ensemble of sinusoidal approaches and co-variance matrix learning-based crossover operator evaluation to increase the capacity of LSHADE-EpSin Comparison of results on IEEE CEC 2017 functions with state-of-the-art algorithms confirmed the enhanced performance SADE-FP Cheng et al [ 66 ] Self-adaptive adjustable parameters were used to make the parameter tuning process easier and a perturbation strategy based on individual fitness performance Comparison with PsDE and canonical DE and other test functions proved its efficiency Db-SHADE Viktorina et al [ 67 ] Alteration was made to mutation and crossover operator selection to eliminate premature convergence in the SHADE algorithm while solving high-dimensional problems The performance of SHADE family algorithms was enhanced using the strategy DEA-06 Greco et al [ 68 ] Used a new scheme for both mutation and crossover in DE. The mutation operator was used adaptively, crossover operator with the replacement of binomial formulation was employed Showed enhanced or competitive results ATBDE Zhao et al [ 69 ] Self-adaptive top–bottom strategy, including failure member scheme with the optional archive Amplified performance of DE or at least comparable to other DE Variants Hard-DE Meng et al [ 70 ] A hierarchical archive-bas...…”