The Human Mental Search Algorithm for Solving Optimisation Problems

Mousavirad, Seyed Jalaleddin; Schaefer, Gerald; Ebrahimpour-Komleh, Hossein

doi:10.1007/978-3-030-52067-0_2

Cited by 10 publications

(1 citation statement)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gradient-based approaches such as back-propagation are widely used for MLNN training, but have drawbacks such as being sensitive to the initial weights and a tendency to get stuck in local optima. Population-based metaheuristic (PBMH) algorithms such as particle swarm optimisation (PSO) [7], differential evolution [8], and human mental search (HMS) [9], [10] are capable of overcoming these problems. PBMHs are problem-independent optimisation algorithms that find an optimal solution using a population of candidate solutions and some specific operators.…”

Section: Introductionmentioning

confidence: 99%

Differential Evolution-based Neural Network Training Incorporating a Centroid-based Strategy and Dynamic Opposition-based Learning

Mousavirad¹,

Oliva²,

Hinojosa³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Training multi-layer neural networks (MLNNs), a challenging task, involves finding appropriate weights and biases. MLNN training is important since the performance of MLNNs is mainly dependent on these network parameters. However, conventional algorithms such as gradient-based methods, while extensively used for MLNN training, suffer from drawbacks such as a tendency to getting stuck in local optima. Population-based metaheuristic algorithms can be used to overcome these problems. In this paper, we propose a novel MLNN training algorithm, CenDE-DOBL, that is based on differential evolution (DE), a centroid-based strategy (Cen-S), and dynamic opposition-based learning (DOBL). The Cen-S approach employs the centroid of the best individuals as a member of population, while other members are updated using standard crossover and mutation operators. This improves exploitation since the new member is obtained based on the best individuals, while the employed DOBL strategy, which uses the opposite of an individual, leads to enhanced exploration. Our extensive experiments compare CenDE-DOBL to 26 conventional and population-based algorithms and confirm it to provide excellent MLNN training performance.

show abstract

Section: Introductionmentioning

confidence: 99%