Random Neural Networks (RNNs) area classof Neural Networks (NNs) that can also be seen as a specific type of queuing network. They have been successfully used in several domains during the last 25 years, as queuing networks to analyze the performance of resource sharing in many engineering areas, as learning tools and in combinatorial optimization, where they are seen as neural systems, and also as models of neurological aspects of living beings. In this article we focus on their learning capabilities, and more specifically, we present a practical guide for using the RNN to solve supervised learning problems. We give a general description of these models using almost indistinctly the terminology of Queuing Theory and the neural one. We present the standard learning procedures usedby RNNs, adapted from similar well-established improvements in the standard NN field. We describe in particular a set of learning algorithms covering techniques based on the use of first order and, then, of second order derivatives. We also discuss some issues related to these objects and present new perspectives about their use in supervised learning problems. The tutorial describes their most relevant applications, and also provides a large bibliography.
The article describes a neural network-based articulatory feature (AF) estimation for the Czech speech. First, the relationship between AFs and a Czech phone inventory is defined, and then the estimation based on the MLP neural networks is done. The usage of several speech representations on the input of the MLP classifiers is proposed with the purpose to obtain a robust AF estimation. The realized experiments have proved that an ANN- based AF estimation works very reliably especially in a low noise environment. Moreover, in case the number of neurons in a hidden layer is increased and if the temporal context DCT-TRAP features are used on the input of the MLP network, the AF classification works accurately also for the signals collected in the environments with a high background noise.
This paper deals with a new approach to designing the micro-electronic system suitable for mass-parallel and neuronal structures realizations in which the high demand on safety and reliability is given. The presented concept is based on the FPGA platform. Authors point out various kinds of faults which can possibly occur during system cycle. Furthermore, authors introduce the Safety Core principle and define systems for which it is applicable. There are possibilities of using partial dynamic reconfiguration shown in this paper in the context of FPGA fabric testing, faults catching and correcting.
By using linear matrix inequality (LMI) approach and Lyapunov functional method, we obtain some new sufficient conditions ensuring global asymptotic stability and global exponential stability of a generalized neutral-type impulsive neural networks with delays. A simulation example is provided to demonstrate the usefulness of the main results obtained. The main contribution in this paper is that a new neutral-type impulsive neural networks with variable delays is studied by constructing a novel Lyapunov functional and LMI approach.
The paper presents new rnethodology liow to decompose the higlh dimensional LTI (linear time invariant) systam with both distinct and repeated eigenvalues of the transition matrix into a set of first-order LTI models, which could be combined to achieve approximation of the original dynamics. As a tool, the Sylvester’s theorems are used to design the filter bank and parameters of the firstorder models (transition values). At the end, the practical examples are shown and the next steps of research of the decomposition theory are indicated.
Backpropagation which uses gradient descent on the steepness of the sigmoid function (BPSA) has been widely studied (e.g. Kruschke et al. [1]). However, most of these studies only analysed the BPSA empirically where no adequate measurements of the network’s quality characteristics (e.g. efficiency and complexity) were given. This paper attempts to show that the BPSA is more efficient than the standard BPA by quantitatively comparing the convergence performance of both algorithms on several benchmark application problems. The convergence performance is measured by the values of the neural metrics [2] evaluated in the training process.
This paper presents a neural network (NN) approach to detect intrusions. Previous works used many KDD records to train NNs for detecting intrusions. That is why; our objective here is to show that in case of the KDD data sets, we can obtain good results by training some NNs with a small data subset. To prove that, this study compares the attacks detection and classification by using two training sets: a set of only 260 records and a set of 65536 records. The testing set is composed of 65536 records randomly chosen from the KDD testing set. Our study focused on two classification types of records: a single class (normal or attack), and a multi class where the category of the attack is detected by the NN. Four different types of NNs were tested: Multi-Layer Perceptron (MLP), Modular, Jordan/Elman and Principal Component Analysis (PCA) NN. Two NN structures were used: the first one contains only one hidden layer and the second contains ten hidden layers. Our simulations show that the small data subset (260 records) can be trained to detect and classify attacks more efficiently than the second data subset.
The purpose of this study is to analyze the performances of some neural networks (NNs) when all the KDD data set is used to train them, in order to classify and detect attacks. Five different types of NNs were tested: Multi-Layer Perceptron (MLP), Self Organization Feature Map (SOFM), Radial Basis Function/Generalized Regression/Probabilistic (RBF/GR/P), Jordan/Elman, and Recurrent NNs. The experiment study is done on the Knowledge Discovery and Data mining (KDD) data sets. We consider two levels of attack granularities depending on whether dealing with four main categories, or only focusing on the normal/attack connection types. Our simulations show that our results are competitive with some other artificial intelligence or data mining intrusion detection systems.
Decrease of attention and an possible micro-sleep of an artificial system operator is very dangerous and its early detection can prevent great losses. This article deals with a classification of States of vigilance based on the analysis of an electroencefalographic activity of the brain. Preprocessing of data is done by the Discrete Fourier Transform. For the recognition radial basis functions (RBF), a k-nearest neighbor and a method based on the Bayesian theory is used. Its coefficients are found using the maximum likelihood estimation. An experiment with recognition of 6 States of vigilance created according to reaction time is performed.