University of Jordan

About: University of Jordan is a education organization based out in Amman, Jordan. It is known for research contribution in the topics: Population & Health care. The organization has 7796 authors who have published 13764 publications receiving 213526 citations.

Training radial basis function networks using biogeography-based optimizer

Abstract: Training artificial neural networks is considered as one of the most challenging machine learning problems. This is mainly due to the presence of a large number of solutions and changes in the search space for different datasets. Conventional training techniques mostly suffer from local optima stagnation and degraded convergence, which make them impractical for datasets with many features. The literature shows that stochastic population-based optimization techniques suit this problem better and are reliably alternative because of high local optima avoidance and flexibility. For the first time, this work proposes a new learning mechanism for radial basis function networks based on biogeography-based optimizer as one of the most well-regarded optimizers in the literature. To prove the efficacy of the proposed methodology, it is employed to solve 12 well-known datasets and compared to 11 current training algorithms including gradient-based and stochastic approaches. The paper considers changing the number of neurons and investigating the performance of algorithms on radial basis function networks with different number of parameters as well. A statistical test is also conducted to judge about the significance of the results. The results show that the biogeography-based optimizer trainer is able to substantially outperform the current training algorithms on all datasets in terms of classification accuracy, speed of convergence, and entrapment in local optima. In addition, the comparison of trainers on radial basis function networks with different neurons size reveal that the biogeography-based optimizer trainer is able to train radial basis function networks with different number of structural parameters effectively.

Ant lion optimizer: Theory, literature review, and application in multi-layer perceptron neural networks

Abstract: This chapter proposes an efficient hybrid training technique (ALOMLP) based on the Ant Lion Optimizer (ALO) to be utilized in dealing with Multi-Layer Perceptrons (MLPs) neural networks. ALO is a well-regarded swarm-based meta-heuristic inspired by the intelligent hunting tricks of antlions in nature. In this chapter, the theoretical backgrounds of ALO are explained in details first. Then, a comprehensive literature review is provided based on recent well-established works from 2015 to 2018. In addition, a convenient encoding scheme is presented and the objective formula is defined, mathematically. The proposed training model based on ALO algorithm is substantiated on sixteen standard datasets. The efficiency of ALO is compared with differential evolution (DE), genetic algorithm (GA), particle swarm optimization (PSO), and population-based incremental learning (PBIL) in terms of best, worst, average, and median accuracies. Furthermore, the convergence propensities are monitored and analyzed for all competitors. The experiments show that the ALOMLP outperforms GA, PBIL, DE, and PSO in classifying the majority of datasets and provides improved accuracy results and convergence rates.

Pharmacophore modeling, quantitative structure-activity relationship analysis, and in silico screening reveal potent glycogen synthase kinase-3beta inhibitory activities for cimetidine, hydroxychloroquine, and gemifloxacin

Abstract: The pharmacophoric space of glycogen synthase kinase-3beta (GSK-3beta) was explored using two diverse sets of inhibitors. Subsequently, genetic algorithm and multiple linear regression analysis were employed to select optimal combination of pharmacophores and physicochemical descriptors that access self-consistent and predictive quantitative structure-activity relationship (QSAR) against 132 training compounds ( r (2) 123 = 0.663, F = 24.6, r (2) LOO = 0.592, r (2) PRESS against 29 external test inhibitors = 0.695). Two orthogonal pharmacophores emerged in the QSAR, suggesting the existence of at least two distinct binding modes accessible to ligands within GSK-3beta binding pocket. The validity of the QSAR equation and the associated pharmacophores was established by the identification of three nanomolar GSK-3beta inhibitors retrieved from our in-house-built structural database of established drugs, namely, hydroxychloroquine, cimetidine, and gemifloxacin. Docking studies supported the binding modes suggested by the pharmacophore/QSAR analysis. In addition to being excellent leads for subsequent optimization, the anti-GSK-3beta activities of these drugs should have significant clinical implications.