Performance Of Neural Model Research Articles

Noninvasive Cuffless Blood Pressure Estimation With Dendritic Neural Regression.

Blood pressure (BP) is one of the most important indicators of health. BP that is too high or too low causes varying degrees of diseases, such as renal impairment, cerebrovascular incidents, and cardiovascular diseases. Since traditional cuff-based BP measurement techniques have the drawbacks of patient discomfort and the impossibility of continuous BP monitoring, noninvasive cuffless continuous BP measurement has become a popular topic. The common noninvasive approach uses machine-learning (ML) algorithms to estimate BP by using the features extracted from simultaneous photoplethysmogram (PPG) and electrocardiogram (ECG) signals, such as the pulse transit time and pulse wave velocity. This study investigates the BP estimation performance of the novel dendritic neural regression (DNR) method proposed by us. Unlike conventional neural networks, DNR utilizes the multiplication operator as the excitation function in each dendritic branch, inspired by biological neuron phenomena, and can effectively capture nonlinear relationships between distinct input features. In addition, AMSGrad is used as the optimization algorithm to further enhance the dendritic neural model's performance. The experimental results show that by being fed a combination of the raw features extracted from the ECG and PPG signals and the components of the BP mathematical models, DNR can increase the accuracy of systolic BP, diastolic BP, and mean arterial pressure estimation significantly, which are superior to the state-of-the-art ML techniques. According to the British Hypertension Society protocol, DNR achieves a grade of A for the long-term BP estimation. Considering its architectural simplicity and powerful performance, the proposed method can be regarded as a reliable tool for estimating long-term continuous BP in a noninvasive cuffless way.

A novel method on the optimization problem of energy conservation in public buildings

ABSTRACT In recent years, the problem of energy consumption has received considerable attention. Among them, the proportion of energy consumption in public buildings accounts for a substantial part, and there is a large space for energy saving in this field. Practically, the major limiting factor for energy consumption optimization is indoor environment comfort degree. As such, a trade-off problem between energy conservation and indoor environment must be resolved. In this paper, an office in a public building is chosen as the experimental object, with 2520 real observations collected for two months, a series of intelligence based techniques are combined and applied to deal with this optimization problem. First, Partial Rank Correlation Coefficient (PRCC) was employed to perform correlation analysis for a set of candidate variables, such that the computational time of subsequent model training is reduced by 8.32 seconds as compared to the case where no PRCC is in use. Following this, Genetic Algorithm (GA)-based Levenberg-Marquardt algorithm (LM) are applied collectively to optimize the conventional Back Propagation (BP) neural network, and results suggest that there was a marked improvement in neural model performance, on one hand the epoch number for convergence drops from 72 to 13, on the other hand the mean squared error for test set is reduced by 7.46e-4 (from 9.93e-4 to 2.47e-4). Finally, pigeon-inspired Optimization (PIO) algorithm is employed to address the formulated optimization problem and the optimal value for Predicted Mean Vote (PMV) and Indoor Air Quality (IAQ) are −0.4775 and 0.007826, respectively. To verify the validity of our results that derived by the proposed approach, corresponding parameters are applied in a practical scenario. The experimental result indicates that the energy efficiency is improved by around 18.6% compared to that yielded before optimization is performed, which demonstrates the applicability and effectiveness of the proposed method.

Modelling of hardfacing layers deposition parameters using robust machine learning algorithms

The study presents a data-driven framework for modelling parameters of hardfacing deposits by GMAW using neural models to estimate the influence of process parameters without the need of creating experimental samples of the material and detailed measurements. The process of GAS Metal Arc Welding (GMAW) hardfacing does sometimes create non-homogenous structures in the material not only in deposited material, but also in the heat-affected zone (HAZ) and base material. Those structures are not fully deterministic, so the modelling method should account for this unpredictable component and only learn the generic structure of the hardness of the resulting material. Artificial neural networks (ANN) were used to create a model of the process using only measured samples without any knowledge of equations governing the process. Robust learning was used to decrease the influence of outliers and noise in the measured data on the neural model performance. The proposed method relies on modification of the loss function and several of them are compared and evaluated as an attempt to construct general framework for analysing the hardness as a function of electric current and arc velocity. The proposed method can create robust models of the hardfacing layers deposition or other welding processes and predict the properties of resulting materials even for unseen parameters based on experimental data. This modelling framework is not typically used for metallurgy, and it requires further case studies to verify its generalisability.

Hierarchical Multi-Granularity Attention- Based Hybrid Neural Network for Text Classification

Neural network-based approaches have become the driven forces for Natural Language Processing (NLP) tasks. Conventionally, there are two mainstream neural architectures for NLP tasks: the recurrent neural network (RNN) and the convolution neural network (ConvNet). RNNs are good at modeling long-term dependencies over input texts, but preclude parallel computation. ConvNets do not have memory capability and it has to model sequential data as un-ordered features. Therefore, ConvNets fail to learn sequential dependencies over the input texts, but it is able to carry out high-efficient parallel computation. As each neural architecture, such as RNN and ConvNets, has its own pro and con, integration of different architectures is assumed to be able to enrich the semantic representation of texts, thus enhance the performance of NLP tasks. However, few investigation explores the reconciliation of these seemingly incompatible architectures. To address this issue, we propose a hybrid architecture based on a novel hierarchical multi-granularity attention mechanism, named Multi-granularity Attention-based Hybrid Neural Network (MahNN). The attention mechanism is to assign different weights to different parts of the input sequence to increase the computation efficiency and performance of neural models. In MahNN, two types of attentions are introduced: the syntactical attention and the semantical attention. The syntactical attention computes the importance of the syntactic elements (such as words or sentence) at the lower symbolic level and the semantical attention is used to compute the importance of the embedded space dimension corresponding to the upper latent semantics. We adopt the text classification as an exemplifying way to illustrate the ability of MahNN to understand texts. The experimental results on a variety of datasets demonstrate that MahNN outperforms most of the state-of-the-arts for text classification.

Multitask learning and benchmarking with clinical time series data

Health care is one of the most exciting frontiers in data mining and machine learning. Successful adoption of electronic health records (EHRs) created an explosion in digital clinical data available for analysis, but progress in machine learning for healthcare research has been difficult to measure because of the absence of publicly available benchmark data sets. To address this problem, we propose four clinical prediction benchmarks using data derived from the publicly available Medical Information Mart for Intensive Care (MIMIC-III) database. These tasks cover a range of clinical problems including modeling risk of mortality, forecasting length of stay, detecting physiologic decline, and phenotype classification. We propose strong linear and neural baselines for all four tasks and evaluate the effect of deep supervision, multitask training and data-specific architectural modifications on the performance of neural models.

Travel time measure specification by functional approximation: application of radial basis function neural networks

In this study, in the purpose of providing a dynamic procedure for reliable travel time specification, the performance of a neural functional approximation method is analysed. The numerical analyses are carried out on the succeeding sections of a freeway segment inputting data obtained from microwave radar sensor units located successively at the cross-sections of a freeway segment of approximately 4km. Measurements on traffic variables, i.e., vehicle counts, speed, and occupancy, for the reference time periods are processed. The structure of the employed radial basis function neural networks are configured considering the data of a three-lane freeway segment obtained by succeeding sensors located in side-fired position. Travel time measures approximated by the neural models are compared with the corresponding field measurements obtained by probe vehicle. Results prove neural model's performance in representing spatiotemporal variation of flow dynamics as well as travel times. Adaptability of the proposed travel time specification procedure to real-time intelligent control systems is a possible future extension.

On the Effects of Signal Acuity in a Multi-Alternative Model of Decision Making

We consider the effects of signal sharpness or acuity on the performance of neural models of decision making. In these models, a vector of signals is presented, and the subject must decide which of the elements of the vector is the largest. McMillen and Holmes ( 2006 ) derived asymptotically optimal tests under the assumption that the elements of the signal vector were all equal except one. In this letter, we consider the case of signals spread around a peak. The acuity is a measure of how strongly peaked the signal is. We find that the optimal test is one in which the detectors are passed through an output layer that encodes knowledge of the possible shapes of the incoming signals. The incorporation of such an output layer can lead to significant improvements in decision-making tasks.

MULTILAYER PERCEPTRON NEURAL ANALYSIS OF EDGE COUPLED AND CONDUCTOR-BACKED EDGE COUPLED COPLANAR WAVEGUIDES

In recent years, Computer Aided Design (CAD) based on Artiflcial Neural Networks (ANNs) have been introduced for microwave modeling, simulation and optimization. In this paper, the characteristic parameters of edge coupled and conductor-backed edge coupled Coplanar Waveguides have been determined with the use of ANN model. Eight learning algorithms, Levenberg-Marquart (LM), Bayesian Regularization (BR), Quasi-Newton (QN), Scaled Conjugate Gradient (SCG), Conjugate Gradient of Fletcher-Powell (CGF), Resilient Propagation (RP), Conjugate Gradient back-propagation with Polak-Ribiere (CGP) and Gradient Descent (GD) are used to train the Multi-Layer Perceptron Neural Networks (MLPNNs). The results of neural models presented in this paper are compared with the results of Conformal Mapping Technique (CMT). The neural results are in very good agreement with the CMT results. When the performances of neural models are compared with each other, the best results are obtained from the neural networks trained by LM and BR algorithms.

An approach to automated classification of epileptic seizures using Artificial Neural Network

Epileptic seizures are important public health issues as they affect 0.8% of humans. Electroencephalograph (EEG) records provide important understanding of epileptic disorders. The conventional method is to interpret by visual inspection of common patterns. This work deals with novel method of data generation using feature extraction and classification using Back Propagation Algorithm. Twelve patients' EEG is used for training and six patients' EEG for testing. Thus, the designed network classifies normal and types of abnormal EEG like focal, absence and tonic-clonic seizures. The network correctly classified normal and abnormal conditions. The performance of neural model has an accuracy of 96.3%.

NEURAL MODELS FOR THE ELLIPTIC- AND CIRCULAR-SHAPED MICROSHIELD LINES

This article presents a new approach based on artificial neural networks (ANNs) to calculate the characteristic parameters of elliptic and circular-shaped microshield lines. Six learning algorithms, bayesian regularization (BR), Levenberg-Marquardt (LM), quasi- Newton (QN), scaled conjugate gradient (SCG), resilient propagation (RP), and conjugate gradient of Fletcher-Reeves (CGF), are used to train the ANNs. The neural results are in very good agreement with the results reported elsewhere. When the performances of neural models are compared with each other, the best and worst results are obtained from the ANNs trained by the BR and CGF algorithms, respectively.

QUASI-STATIC MODELS BASED ON ARTIFICIAL NEURAL NEWORKS FOR CALCULATING THE CHARACTERISTIC PARAMETERS OF MULTILAYER CYLINDRICAL COPLANAR WAVEGUIDE AND STRIP LINE

In this paper, two different neural models are proposed for calculating the quasi-static parameters of multilayer cylindrical coplanar waveguides and strip lines. These models were basically developed by training the artificial neural networks with the numerical results of quasi-static analysis. Neural models were trained with four different learning algorithms to obtain better performance and faster convergence with simpler structure. When the performances of neural models are compared with each other, the best test results are obtained from the multilayered perceptrons trained by the Levenberg- Marquardt algorithm. The results obtained from the neural models are in very good agreements with the theoretical results available in the literature.

NEURAL MODELS FOR THE BROADSIDE-COUPLED V-SHAPED MICROSHIELD COPLANAR WAVEGUIDES

This article presents a new approach based on multilayered perceptron neural networks (MLPNNs) to calculate the odd-and even-mode characteristic impedances and effective permittivities of the broadside-coupled V-shaped microshield coplanar waveguides (BC-VSMCPWs). Six learning algorithms, bayesian regulation (BR), Levenberg-Marquardt (LM), quasi-Newton (QN), scaled conjugate gradient (SCG), resilient propagation (RP), and conjugate gradient of Fletcher-Powell (CGF), are used to train the MLPNNs. The neural results are in very good agreement with the results reported elsewhere. When the performances of neural models are compared with each other, the best and worst results are obtained from the MLPNNs trained by the BR and CGF algorithms, respectively.

A comparative study of neural networks for input resistance computation of electrically thin and thick rectangular microstrip antennas

Multilayered-perceptron-based neural models for calculating the input resistance of electrically thin and thick rectangular microstrip antennas are presented. Eleven learning algorithms—Levenberg-Marquardt, conjugate gradient of Fletcher-Reeves, Bayesian regularization, Broyden-Fletcher-Goldfarb-Shanno, conjugate gradient of Polak-Ribiere, conjugate gradient of Powell-Beale, scaled conjugate gradient, one-step secant, resilient backpropagation, backpropagation with momentum, and backpropagation with adaptive learning rate—are used to train the multilayered perceptrons. The input resistance results obtained with the use of neural models are in very good agreement with the experimental results available in the literature. When the performances of neural models are compared with each other, the best result is obtained from the multilayered perceptrons trained by the Levenberg-Marquardt algorithm.

Neural models for the V‐shaped conductor‐backed coplanar waveguides

AbstractNeural models based on the artificial neural networks (ANNs) for computing the effective permittivities and characteristic impedances of V‐shaped conductor‐backed coplanar waveguides are presented. The proposed neural models can also be used for calculating the characteristic parameters of conductor‐backed coplanar waveguides. Six learning algorithms, Levenberg–Marquardt, bayesian regularization, quasi–Newton, scaled conjugate gradient, conjugate gradient of Fletcher‐Reeves, and resilient propagation, are used to train the ANNs. The neural results are in very good agreement with the results available in the literature. When the performances of neural models are compared with each other, the best result is obtained from the ANNs trained by the LM algorithm. © 2007 Wiley Periodicals, Inc. Microwave Opt Technol Lett 49: 1294–1299, 2007; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mop.22473

Neural models for quasi-static analysis of conventional and supported coplanar waveguides

Neural models based on multilayered perceptrons (MLPs) for computing the effective permittivities and the characteristic impedances of both the conventional coplanar waveguides (CCPWs) and the supported coplanar waveguides (SCPWs) are presented. Six learning algorithms, Levenberg–Marquardt (LM), Bayesian regularization (BR), quasi-Newton (QN), scaled conjugate gradient (SCG), conjugate gradient of Fletcher–Powell (CGF), and resilient propagation (RP), are used to train the MLPs. The results of neural models presented in this paper are compared with the results of the experimental works, the conformal mapping technique (CMT), the spectral domain approach (SDA), and three commercial electromagnetic simulators such as IE3D, CAPIND2D, and MMICTL. The neural results are in very good agreement with the theoretical and experimental results. When the performances of neural models are compared with each other, the best result is obtained from the MLPs trained by the LM algorithm.

Artificial Neural Networks for Calculating the Characteristic Impedance of Air-Suspended Trapezoidal and Rectangular-Shaped Microshield Lines

Neural models for calculating the characteristic impedance of air-suspended trapezoidal and rectangular-shaped microshield lines, based on the multilayered perceptrons (MLPs), are presented. Six learning algorithms, bayesian regulation (BR), Levenberg-Marquardt (LM), quasi-Newton (QN), scaled conjugate gradient (SCG), resilient propagation (RP), and conjugate gradient of Fletcher-Powell (CGF), are used to train the MLPs. The characteristic impedance results obtained by using neural models are in very good agreement with the results available in the literature. When the performances of neural models are compared with each other, the best test result is obtained from the MLPs trained by the BR algorithm.

Artificial Neural Networks for Resonant Frequency Calculation of Rectangular Microstrip Antennas with Thin and Thick Substrates

Neural models based on multilayered perceptrons for computing the resonant frequency of rectangular microstrip antennas with thin and thick substrates are presented. Eleven learning algorithms, Levenberg-Marquardt, conjugate gradient of Fletcher-Reeves, conjugate gradient of Powell-Beale, bayesian regularization, scaled conjugate gradient, Broyden-Fletcher-Goldfarb-Shanno, resilient backpropagation, conjugate of Polak-Ribiere, backpropagation with adaptive learning rate, one-step secant, and backpropagation with momentum, are used to train the multilayered perceptrons. The resonant frequency results obtained by using neural models are in very good agreement with the experimental results available in the literature. When the performances of neural models are compared with each other, the best result is obtained from the multilayered perceptrons trained by Levenberg-Marquardt algorithm.

Neural Models for the Resonant Frequency of Electrically Thin and Thick Circular Microstrip Antennas and the Characteristic Parameters of Asymmetric Coplanar Waveguides Backed with a Conductor

Summary Neural models for computing the resonant frequency of electrically thin and thick circular microstrip antennas, based on the multilayered perceptrons and the radial basis function networks, are presented. Five learning algorithms, delta-bar-delta, extended delta-bar-delta, quick-propagation, directed random search and genetic algorithms, are used to train the multilayered perceptrons. The radial basis function network is trained according to its learning strategy. The resonant frequency results of neural models are in very good agreement with the experimental results available in the literature. In this paper, the characteristic impedance and the effective permittivity of the asymmetric coplanar waveguide backed with a conductor are also computed by using only one neural model trained by the backpropagation with momentum and the extended delta-bar-delta algorithms. When the performances of neural models are compared with each other, the best results for test are obtained from the multilayered perceptrons trained by the extended delta-bar-delta algorithm.

Comparison of Neural Networks for Resonant Frequency Computation of Electrically Thin and Thick Rectangular Microstrip Antennas

Neural models for calculating the resonant frequency of electrically thin and thick rectangular microstrip antennas, based on the multilayered perceptrons and the radial basis function networks, are presented. Six learning algorithms, backpropagation, delta-bar-delta, extended-delta-bar-delta, quick-propagation, directed random search and genetic algorithms, are used to train the multilayered perceptrons. The radial basis function network is trained according to its learning strategy. The reason for using six different learning algorithms and two different structures is to speed up the training time and to compare the performance of neural models for this specific application. The resonant frequency results obtained by using neural models are in very good agreement with the experimental results available in the literature. When the performances of neural models are compared with each other, the best results for training and test were obtained from the radial basis function network