Dynamic scaling factor based differential evolution with multi-layer perceptron for gene selection from pathway information of microarray data

Quantum-Inspired Differential Evolution with Decoding using Hashing for Efficient User Allocation in Edge Computing Environment

“…With the optimized parameters (Table VIII), majority of the strategies (QDE-UA, QPSO [8], and DE [11]) perform well by providing optimal Γ. This shows that the number of obtained ESs is ideal for allocating the AUs as in Table VIII.…”

“…Differential evolution (DE) is a well-known evolutionary algorithm for solving complex optimization problems [11], [18]. DE usually consists of four steps: vector initialization, mutation, crossover, and selection.…”

“…Here QDE-UA is compared with an existing user allocation strategy, EUA-FOA [10], along with other metaheuristic algorithms such as DE [11] and QPSO [8]. The population size is taken within the range of 150-250, the maximum iteration is taken as 3000-4000, the scaling factor (F ) of DE is considered as 0.01 ≤ F ≤ 0.005, and the crossover rate is 0.72.…”

“…In scenario 1 (Fig. 8(b)), the Γ of QPSO [8], and QDE-UA are significantly better than EUA-FOA [10] and DE [11]. This implies that quantum-inspired strategies have better capabilities of finding optimal solutions.…”

“…• Design of experiment (DoE) is performed using the Taguchi method [8], [9]. Extensive simulations are performed and the proposed QDE-UA is compared with the existing strategies such as EUA-FOA [10], DE [11], and QPSO [8]. Hypothesisbased statistical analyses (ANOVA, Friedman test) are also conducted to illustrate the significance of the proposed QDE-UA.…”