Mean square exponential stability of stochastic Cohen-Grossberg neural networks with unbounded distributed delays.
Mixture of experts architectures for neural networks as a special case of conditional expectation formula
Recently a new interesting architecture of neural networks called “mixture of experts” has been proposed as a tool of real multivariate approximation or prediction. We show that the underlying problem is closely related to approximating the joint probability density of involved variables by finite mixture. Particularly, assuming normal mixtures, we can explicitly write the conditional expectation formula which can be interpreted as a mixture-of- experts network. In this way the related optimization...
Modelling of Cancer Growth, Evolution and Invasion: Bridging Scales and Models
Since cancer is a complex phenomenon that incorporates events occurring on different length and time scales, therefore multiscale models are needed if we hope to adequately address cancer specific questions. In this paper we present three different multiscale individual-cell-based models, each motivated by cancer-related problems emerging from each of the spatial scales: extracellular, cellular or subcellular, but also incorporating relevant information from other levels. We apply these hybrid...
Motor control neural models and systems theory
In this paper, we introduce several system theoretic problems brought forward by recent studies on neural models of motor control. We focus our attention on three topics: (i) the cerebellum and adaptive control, (ii) reinforcement learning and the basal ganglia, and (iii) modular control with multiple models. We discuss these subjects from both neuroscience and systems theory viewpoints with the aim of promoting interplay between the two research communities.
Multiple neural network integration using a binary decision tree to improve the ECG signal recognition accuracy
The paper presents a new system for ECG (ElectroCardioGraphy) signal recognition using different neural classifiers and a binary decision tree to provide one more processing stage to give the final recognition result. As the base classifiers, the three classical neural models, i.e., the MLP (Multi Layer Perceptron), modified TSK (Takagi-Sugeno-Kang) and the SVM (Support Vector Machine), will be applied. The coefficients in ECG signal decomposition using Hermite basis functions and the peak-to-peak...
Neural network based feedback linearization control of a servo-hydraulic vehicle suspension system
This paper presents the design of a neural network based feedback linearization (NNFBL) controller for a two degree-offreedom (DOF), quarter-car, servo-hydraulic vehicle suspension system. The main objective of the direct adaptive NNFBL controller is to improve the system's ride comfort and handling quality. A feedforward, multi-layer perceptron (MLP) neural network (NN) model that is well suited for control by discrete input-output linearization (NNIOL) is developed using input-output data sets...
Neural network based identification of hysteresis in human meridian systems
Developing a model based digital human meridian system is one of the interesting ways of understanding and improving acupuncture treatment, safety analysis for acupuncture operation, doctor training, or treatment scheme evaluation. In accomplishing this task, how to construct a proper model to describe the behavior of human meridian systems is one of the very important issues. From experiments, it has been found that the hysteresis phenomenon occurs in the relations between stimulation input and...
Neural network segmentation of images from stained cucurbits leaves with colour symptoms of biotic and abiotic stresses
The increased production of Reactive Oxygen Species (ROS) in plant leaf tissues is a hallmark of a plant's reaction to various environmental stresses. This paper describes an automatic segmentation method for scanned images of cucurbits leaves stained to visualise ROS accumulation sites featured by specific colour hues and intensities. The leaves placed separately in the scanner view field on a colour background are extracted by thresholding in the RGB colour space, then cleaned from petioles to...
Neural networks with memory.
Neuromorphic features of probabilistic neural networks
We summarize the main results on probabilistic neural networks recently published in a series of papers. Considering the framework of statistical pattern recognition we assume approximation of class-conditional distributions by finite mixtures of product components. The probabilistic neurons correspond to mixture components and can be interpreted in neurophysiological terms. In this way we can find possible theoretical background of the functional properties of neurons. For example, the general...
Neurons: a numerical approach.
New improved exponential stability criteria for discrete-time neural networks with time-varying delay.
New result of existence of periodic solution for a Hopfield neural networks with neutral time-varying delays.
New results of a class of two-neuron networks with time-varying delays.
New sufficient conditions for global asymptotic stability of a kind of nonlinear neutral differential equations
This paper addresses the stability study for nonlinear neutral differential equations. Thanks to a new technique based on the fixed point theory, we find some new sufficient conditions ensuring the global asymptotic stability of the solution. In this work we extend and improve some related results presented in recent works of literature. Two examples are exhibited to show the effectiveness and advantage of the results proved.
Noise Shaping in Neural Populations with Global Delayed Feedback
The interplay between intrinsic and network dynamics has been the focus of many investigations. Here we use a combination of theoretical and numerical approaches to study the effects of delayed global feedback on the information transmission properties of neural networks. Specifically, we compare networks of neurons that display intrinsic interspike interval correlations (nonrenewal) to networks that do not (renewal). We find that excitatory and...
Nonlinear predictive control based on neural multi-models
This paper discusses neural multi-models based on Multi Layer Perceptron (MLP) networks and a computationally efficient nonlinear Model Predictive Control (MPC) algorithm which uses such models. Thanks to the nature of the model it calculates future predictions without using previous predictions. This means that, unlike the classical Nonlinear Auto Regressive with eXternal input (NARX) model, the multi-model is not used recurrently in MPC, and the prediction error is not propagated. In order to...
Nonparametric statistical analysis for multiple comparison of machine learning regression algorithms
In the paper we present some guidelines for the application of nonparametric statistical tests and post-hoc procedures devised to perform multiple comparisons of machine learning algorithms. We emphasize that it is necessary to distinguish between pairwise and multiple comparison tests. We show that the pairwise Wilcoxon test, when employed to multiple comparisons, will lead to overoptimistic conclusions. We carry out intensive normality examination employing ten different tests showing that the...
On classification with missing data using rough-neuro-fuzzy systems
The paper presents a new approach to fuzzy classification in the case of missing data. Rough-fuzzy sets are incorporated into logical type neuro-fuzzy structures and a rough-neuro-fuzzy classifier is derived. Theorems which allow determining the structure of the rough-neuro-fuzzy classifier are given. Several experiments illustrating the performance of the roughneuro-fuzzy classifier working in the case of missing features are described.
On global exponential stability of discrete-time Hopfield neural networks with variable delays.