Most Artificial Neural Networks (ANNs) have a fixed topology during learning, and typically suffer from a number of shortcomings as a result. Variations of ANNs that use dynamic topologies have shown ability to overcome many of these problems. This paper introduces Location-Independent Transformations (LITs) as a general strategy for implementing neural networks that use static and dynamic topo...

Artificial Neural Networks (ANNs) have been used to perform classification for Automatic Speech Recognition (ASR). In this paper, we propose a new neural network, the Contenders' Network (CN) which requires little initial knowledge of the classification problem and lesser neurons than other ANNS

Artificial neural network was considered in previous studies for prediction of engine performance and emissions. ICA methodology was inspired in order to optimize the weights of multilayer perceptron (MLP) of artificial neural network so that closer estimation of output results can be achieved. Current paper aimed at prediction of engine power, soot, NOx, CO2, O2, and temperature with the ai...

In this paper a competitive neural network with binary synaptic weights is proposed. The aim of this network is to cluster or categorize binary input data. The neural network uses a learning mechanism based on activity levels that generates new binary synaptic weights that become medianoids of the clusters or categorizes that are being formed by process units of the network, since the medianoid...

An unsupervised competitive learning algorithm based on the classical -means clustering algorithm is proposed. The proposed learning algorithm called the centroid neural network (CNN) estimates centroids of the related cluster groups in training date. This paper also explains algorithmic relationships among the CNN and some of the conventional unsupervised competitive learning algorithms includ...

This paper describes a practical segmentation procedure using a simple competitive learning neural network to yield a complete segmentation suitable for segmentation-based image coding. Image segmentation here is considered as a vector quantization problem. The procedure using the FSCL neural network for the vector quantization has the two main parts, that is, primary and secondary segmentation...

This paper presents the problem of multiple quadrature amplitude modulated signals equalization and argues the use of a radial basis functions neural network (RBF-NN) equalizer. Different competitive learning algorithms for the RBF-NN centres determination are discussed. A new competitive learning algorithm is introduced, the rival penalized competitive learning, which rewards the winner and pe...

Ensemble approaches have been shown to enhance classification by combining the outputs from a set of voting classifiers. Diversity in error patterns among base classifiers promotes ensemble performance. Multi-task learning is an important characteristic for Neural Network classifiers. Introducing a secondary output unit that receives different training signals for base networks in an ensemble c...