Global exponential stability of delayed competitive neural networks with different time scales

The dynamics of cortical cognitive maps developed by selforganization must include the aspects of long and short-term memory. The behavior of such a neural network is characterized by an equation of neural activity as a fast phenomenon and an equation of synaptic modification as a slow part of the neural system, besides, this model bases on unsupervised synaptic learning algorithm. In this paper, using theory of the topological degree and strict Liapunov functional methods, we prove existence and uniqueness of the equilibrium of competitive neural networks with continuously distributed delays and different time scales, and present some new criteria for its global robust stability.

Information processing and two types of memory in an analog neural network model with time delay that produces chaos similar to the human and animal EEGs are considered. There are two levels of information processing in this neural network: the level of individual neurons and the level of the neural network. Similar to the state of brain, the state of chaotic neural network is defined. It is characterized by two types of memories (memory I and memory II) and correlation structure between the neurons. In normal (unperturbed) state, the neural network generates chaotic patterns of averaged neuronal activities (memory I) and patterns of oscillation amplitudes (memory II). In the presence of external stimulation, the activity patterns change, showing changes in both types of memory. As in experiments on stimulation of the brain, the neural network model shows synchronization of neuronal activities due to stimulus measured by Pearson's correlation coefficient. An increase in neural network asymmetry (increase of the neural network excitability) leads to the phenomenon similar to the epilepsy. Modeling of brain injury, Parkinson's disease, and dementia is performed by removing and weakening interneuron connections. In all cases, the chaotic neural network shows a decrease of the degree of chaos and changes in both types of memory similar to those observed in experiments with healthy human subjects and patients with Parkinson's disease and dementia. © 2005 Wiley Periodicals, Inc. Complexity 11:39–52, 2005

The Sixth International Symposium on Neural Networks (ISNN 2009)

1. In this study, neural network (NN) and response surface (RS) models were developed to investigate the response [average daily gain (ADG) and feed efficiency (FE)] of young broiler chickens to dietary protein and lysine. For this purpose, data on their responses to dietary protein and lysine were extracted from the literature and separate NN and RS models were constructed. 2. Comparison between the NN and RS models revealed higher accuracy of prediction with the NN models compared to the RS models. In terms of R 2 values, the NN models developed for both ADG (R 2 = 0.923) and FE (R 2 = 0.904) were far superior to the RS models (R 2 for ADG = 0.511; R 2 for FE = 0.67). This suggests that the NN models can serve as an alternative option to conventional regression approaches including use of RS models. 3. Optimisation of the NN models developed for response to protein and lysine showed that diets containing 220.7 (g/kg of diet) protein and 12.85 (g/kg of diet) lysine maximise ADG, whereas maximum FE is achieved with diets containing 241.3 and 13.12 (g/kg) protein and lysine, respectively. Based on the optimisation results, optimal dietary protein and lysine concentrations for maximum FE in broiler chickens during the starting period are higher than for ADG.

Background: Data mining (DM) is an approach used in extracting valuable information from environmental processes. This research depicts a DM approach used in extracting some information from influent and effluent wastewater characteristic data of a waste stabilization pond (WSP) in Birjand, a city in Eastern Iran. Methods: Multiple regression (MR) and neural network (NN) models were examined using influent characteristics (pH, Biochemical oxygen demand [BOD 5 ], temperature, chemical oxygen demand [COD], total suspended solids [TSS], total dissolved solid [TDS], electrical conductivity [EC] and turbidity) as the regression input vectors. Models were adjusted to input attributes, effluent BOD 5 (BODout) and COD (CODout). The models performances were estimated by 10-fold external cross-validation. An internal 5-fold cross-validation was also used for the training data set in NN model. The models were compared using regression error characteristic (REC) plot and other statistical measures such as relative absolute error (RAE). Sensitivity analysis was also applied to extract useful knowledge from NN model. Results: NN models (with RAE = 78.71 ± 1.16 for BODout and 83.67 ± 1.35 for CODout) and MR models (with RAE = 84.40% ± 1.07 for BODout and 88.07 ± 0.80 for CODout) indicate different performances and the former was better (P < 0.05) for the prediction of both effluent BOD 5 and COD parameters. For the prediction of CODout the NN model with hidden layer size (H) = 4 and decay factor = 0.75 ± 0.03 presented the best predictive results. For BODout the H and decay factor were found to be 4 and 0.73 ± 0.03, respectively. TDS was found as the most descriptive influent wastewater characteristics for the prediction of the WSP performance. The REC plots confirmed the NN model performance superiority for both BOD and COD effluent prediction. Conclusion: Modeling the performance of WSP systems using NN models along with sensitivity analysis can offer better understanding on exploring the most significant parameters for the prediction of system performance. The findings of this study could build the foundation for prospective work on the characterization of WSP operations and optimization of their performances with a view to conducting statistical approaches.

Global stability analysis of competitive neural networks with mixed time-varying delays and discontinuous neuron activations

This study investigates the application of advanced deep learning models to forecast fossil energy prices, a critical factor influencing global economic stability. Unlike previous research, this study conducts a comparative analysis of Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Bidirectional Long Short-Term Memory (Bi-LSTM), Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and Deep Neural Network (DNN) models. The evaluation metrics employed include Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE). The results reveal that recurrent architectures, particularly GRU, LSTM, and Bi-LSTM, consistently outperform feedforward and convolutional models, demonstrating superior ability to capture temporal dependencies and nonlinear dynamics in energy markets. In contrast, the RNN and DNN show relatively weaker generalization capabilities. Additionally, visualizations of actual versus predicted prices for each model further emphasize superior forecasting accuracy of recurrent models. The results highlight the potential of deep learning in enhancing investment and policy decisions. Additionally, the results provide significant implications for policymakers and investors by emphasizing the value of accurate energy price forecasting in mitigating market volatility, improving portfolio management, and supporting evidence-based energy policies.

Long short-term memory - Fully connected (LSTM-FC) neural network for PM2.5 concentration prediction

The study presents the results of the development and testing of deep learning neural network architectures, which demonstrate high accuracy rates in classifying neurophysiological data, in particular hemodynamic brain activation patterns obtained by functional near-infrared spectroscopy, during solving mathematical problems on spatial-numerical associations. The analyzed signal represents a multidimensional time series of oxyhemoglobin and deoxyhemoglobin dynamics. Taking the specificity of the fNIRS signal into account, a comparative analysis of 2 types of neural network architectures was carried out: (1) architectures based on recurrent neural networks: recurrent neural network with long short-term memory, recurrent neural network with long short-term memory with fully connected layers, bidirectional recurrent neural network with long short-term memory, convolutional recurrent neural network with long short-term memory; (2) architectures based on convolutional neural networks with 1D convolutions: convolutional neural network, fully convolutional neural network, residual neural network. Trained long short-term memory recurrent neural network architectures showed worse results in accuracy in comparison with 1D convolutional neural network architectures. Residual neural network (model_Resnet) demonstrated the highest accuracy rates in three experimental conditions more than 88% in detecting age-related differences in brain activation during spatial-numerical association tasks considering the individual characteristics of the respondents’ signal.

This paper presents an application of the neural network (NN) model for forecasting and filtering problems. Instead of using the gradient descent method as used by the standard back-propagation network, the present study adopted the Levenberg-Marquardt algorithm. It was shown that the Levenberg-Marquardt algorithm tends to reach termination criteria more quickly making the searches employing the full model less time consuming. In the first part of forecasting problem, multilayer feedforward (MLFF) network was constructed by trials to forecast the tidal-level variations at the Chao Phraya river mouth in Thailand. Unlike the well-known conventional harmonic analysis, the neural network model uses a set of previous data for learning and then forecasting directly the time series of tidal levels. It was found that lead time of 1 to 24 hourly tidal levels can be successfully predicted using only a short-time hourly learning data.In the second part of filtering problem, the MLFF network was also used to establish a stage-discharge relationship for tidal river. Comparisons were made among conventional methods, i.e. by using multivariate functions (Linear and power models) and the NN model. The NN model was found to be able to model the transfer function too much higher accuracy than multiple regression analysis. The root mean square errors (RMSE) and the mean relative error (MRE) for the NN model are about 8 cm and 6.4% whereas the same errors for multiple regression analysis are 16 cm and 12%, respectively. The efficiency index for the linear, power, and NN models is 0.4, 0.7, and 0.9, respectively. In addition, the obtained stage-discharge relationship by the NN model can indicate reasonably important behavior of the tidal influences, i.e. there are stronger influences of tide during the dry period than the flood period.

We propose a class of neutral type quaternion-valued neural networks with delays in the leakage term on time scales that can unify the discrete-time and the continuous-time neural networks. In order to avoid the difficulty brought by the noncommutativity of quaternion multiplication, we first decompose the quaternion-valued system into four real-valued systems. Then, by applying the exponential dichotomic theory of linear dynamic equations on time scales, Banach’s fixed point theorem, the theory of calculus on time scales, and inequality techniques, we obtain some sufficient conditions on the existence and global exponential stability of pseudo almost periodic solutions for this class of neural networks. Our results are completely new even for both the case of the neural networks governed by differential equations and the case of the neural networks governed by difference equations and show that, under a simple condition, the continuous-time quaternion-valued network and its corresponding discrete-time quaternion-valued network have the same dynamical behavior for the pseudo almost periodicity. Finally, a numerical example is given to illustrate the feasibility of our results.

This paper proposes the basic structure of a competitive neural network (CNN), based on analog IC techniques, for hardware implementation. This leads to a more compact project and allows real time processing. It is shown that the fundamental equations that describe the behavior of competitive neural networks possess a relationship with some basic electronic components. This fact allows the direct implementation of CNN with these electronic components. Initially the behavior of the fundamental equations of this type of neural networks is studied by means of software simulations. This behavior is then compared with the one obtained through electric simulations of the equivalent circuits originated from these fundamental equations. Simulations show that the most important features of the CNN are obtained with the presented implementation. Finally, a typical application is presented in the area of pattern clustering using synaptic weights to demonstrate an implementation using the techniques described.

Existence and global exponential stability of a periodic solution to interval general bidirectional associative memory (BAM) neural networks with multiple delays on time scales

To improve coastal adaptation and management, it is critical to better understand and predict the characteristics of sea levels. Here, we explore the capabilities of artificial intelligence, from four deep learning methods to predict the surge component of sea-level variability based on local atmospheric conditions. We use an Artificial Neural Networks, Convolutional Neural Network, Long Short-Term Memory layer (LSTM) and a combination of the latter two (ConvLSTM), to construct ensembles of Neural Network (NN) models at 736 tide stations globally. The NN models show similar patterns of performance, with much higher skill in the mid-latitudes. Using our global model settings, the LSTM generally outperforms the other NN models. Furthermore, for 15 stations we assess the influence of adding complexity more predictor variables. This generally improves model performance but leads to substantial increases in computation time. The improvement in performance remains insufficient to fully capture observed dynamics in some regions. For example, in the tropics only modelling surges is insufficient to capture intra-annual sea level variability. While we focus on minimising mean absolute error for the full time series, the NN models presented here could be adapted for use in forecasting extreme sea levels or emergency response.

Solar desalination is one of the renewable energy techniques by which freshwater can be obtained economically. Solar desalination experiments are time and resource-consuming methods; hence there is a need for a robust system to identify the serviceability of the solar still in a specific region. The objective of the present study is to develop a forecasting model using artificial neural networks to predict freshwater productivity. Specifically, the study aims to compare the accuracy of Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) neural networks in forecasting the productivity of Conventional Solar Stills (CSS) and Solar Stills with Phase Change Material (SSPCM). Additionally, the research seeks to analyze the impact of advanced materials, such as phase change materials (PCM) and pin fins, on solar still productivity; determine which neural network model provides more accurate and reliable forecasts for different solar still designs; and contribute to the optimization of solar still implementations by developing predictive models applicable to diverse geographical regions. The current investigation involved analysing the experimental outcomes of a solar still that employed phase change material (PCM) and pin fins. Palmitic acid was implemented as the energy storage material and was placed beneath the absorber plate. The neural network model was trained and validated using time-series solar still experimental data. Different statistical measures were utilised to evaluate the accuracy of Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). The results indicate that the freshwater productivity forecasted by LSTM exhibited greater accuracy than GRU. Specifically, the coefficient of determination values for LSTM were 0.96 and 0.98 for the CSS and SSPCM, respectively, which were higher than the corresponding values for GRU.