New learning strategy for supervised neural network: MPCA meta-heuristic approach

Abstract

The problem of parameter optimization for a feedforward artificial neural network (ANN) to determined its best architecture is addressed. A new metaheuristic called Multiple Particle Collision Algorithm (MPCA), introduced by Luz et al. [12], was applied to design an optimum architecture for two models of supervised neural network: the Multilayer Perceptron (MLP), and recurrent Elman network. The NN obtained using this approach is said to be self-configurable. In addition, two strategies are employed for calculating the connection weights to the MLP and Elman networks: MPCA, and backpropagation algorithm. The resulting ANNs were applied to predict the monthly mesoscale climate for the precipitation field. The comparison is performed between the ANN configuration obtained by automatic process and another configuration proposed by a human specialist.

Model Based Demand Order Estimation by Using Optimal Architecture Artificial Neural Network with Metaheuristic Optimizations

“…The main purpose of automatically configuring an ANN model is the ability to obtain a nearoptimal ANN architecture without requiring the ANN approach and/or the knowledge of any expert in its implementation (Anochi et al, 2014;Anochi et al, 2016). The results obtained in this way prevent unnecessary loss of time and effort.…”

“…Architectural optimization is only performed by using metaheuristic algorithms. This objective function was previously improved for the self-configuring of ANNs by using a multi-particle collision algorithm in (Anochi et al, 2016;Badr et al, 2019). In the current study, Carvalho et al's objective is enhanced to perform more effective heuristic search of neural architecture parameters.…”

Optimal architecture artificial neural network model design with exploitative alpha gray wolf optimization for soft calibration of CO concentration measurements in electronic nose applications

“…The main advantage of using an automated procedure (self-design procedure) to configure an ANN is the ability to obtain a near-optimal ANN architecture without the need for manual trial and error on ANN configurations or the help of an expert in design of ANNs (Anochi et al, 2013(Anochi et al, , 2015. Such a self-configuration approach avoids time-consuming manual trial and error cycles to find out the optimal architecture of neural network models.…”