In the eighties the problem of the lack of an efficient algorithm to train multilayer Rosenblatt perceptrons was solved by sigmoidal neural networks and backpropagation. But should we still try to find an efficient algorithm to train multilayer hardlimit neuronal networks, a task known as a NP-Complete problem? In this work we show that this would not be a waste of time by means of a counter ex...

Dynamic neural network (DNN) models provide an excellent means for forecasting and prediction of nonstationary time series. A neural network architecture, known as locally recurrent neural network ((LRNN) [71], is preferred to the traditional multilayer perceptron (MLP) because the time varying nature of a stock time series can be better represented using LRNN. The use of LRNN has demonstrated ...

In this paper, Levenberg-Marquardt (LM) learning algorithm for a single Integrate-and-Fire Neuron (IFN) is proposed and tested for various applications in which a neural network based on multilayer perceptron is conventionally used. It is found that a single IFN is sufficient for the applications that require a number of neurons in different hidden layers of a conventional neural network. Sever...

Sensor technology has been used in water environment, which comes into being a water environment wireless sensor monitoring network. Monitoring data in the network slowly change, so we propose a geographical energy-efficient multi-hop clustering fusion routing algorithm based on multilayer perceptron (MLP-GEEMHCFR) in this paper to reduce transmittingdata and save the network energy.The algorit...

The presented paper compares forecast of drought indices based on two different models of artificial neural networks. The first model is based on feedforward multilayer perceptron, sANN, and the second one is the integrated neural network model, hANN. The analyzed drought indices are the standardized precipitation index (SPI) and the standardized precipitation evaporation index (SPEI) and were ...

In this work, generalization ability of a hybrid neural network algorithm is investigated. This algorithm consists of a combination of Radial Basis Function (RBF) and Multilayer Perceptron (MLP) in one single network using conic section functions. The network architecture using this algorithm is called Conic Section Function Neural Network (CSFNN). Various problems are examined to demonstrate t...

One of the central issues in neural network research is how to find an optimal MultiLayer Perceptron architecture. The number of neurons, their organization in layers, as well as their connection scheme have a considerable influence on network learning, and on the capacity for generalization [7]. A solution to find out these parameters is needed: The neuro-evolution ([1,2,4,5]). The novelty is ...

In this paper we present our teamwork main results in the area of the automatic speech recognition and understanding in Romanian language using hidden Markov models (HMM), artificial neural networks (multilayer perceptron, support vector machines, Kohonen networks) and hybrid models (fuzzy HMM, fuzzy multilayer perceptron, HMM/multilayer perceptron) for different small Romanian language corpora...

nowadays, software cost estimation (sce) with machine learning techniques are more performance than other traditional techniques which were based on algorithmic techniques. in this paper, we present a new hybrid model of multi-layer perceptron (mlp) artificial neural network (ann) and ant colony optimization (aco) algorithm for high accuracy in sce called multilayer perceptron ant colony optimi...