Search

Recent increases in global food demand have made this research and, therefore, the prediction of agricultural commodity prices, almost imperative. The aim of this paper is to build efficient artificial intelligence methods to effectively forecast commodity prices in light of these global events. Using three separate, well-structured models, the commodity prices of eleven major agricultural commodities that have recently caused crises around the world have been predicted. In achieving its objective, this paper proposes a novel forecasting model for agricultural commodity prices using the extreme learning machine technique optimized with the genetic algorithm. In predicting the eleven commodities, the proposed model, the extreme learning machine with the genetic algorithm, outperforms the model formed by the combination of long short-term memory with the genetic algorithm and the autoregressive integrated moving average model. Despite the fluctuations and changes in agricultural commodity prices in 2022, the extreme learning machine with the genetic algorithm model described in this study successfully predicts both qualitative and quantitative behavior in such a large number of commodities and over such a long period of time for the first time. It is expected that these predictions will provide benefits for the effective management, direction and, if necessary, restructuring of agricultural policies by providing food requirements that adapt to the dynamic structure of the countries. [ABSTRACT FROM AUTHOR]

Once realized, autonomous aerial refueling will revolutionize unmanned aviation by removing current range and endurance limitations. Previous attempts at establishing vision-based solutions have come close but rely heavily on near perfect extrinsic camera calibrations that often change midflight. In this paper, we propose dual object detection, a technique that overcomes such requirement by transforming aerial refueling imagery directly into receiver aircraft reference frame probe-to-drogue vectors regardless of camera position and orientation. These vectors are precisely what autonomous agents need to successfully maneuver the tanker and receiver aircraft in synchronous flight during refueling operations. Our method follows a common 4-stage process of capturing an image, finding 2D points in the image, matching those points to 3D object features, and analytically solving for the object pose. However, we extend this pipeline by simultaneously performing these operations across two objects instead of one using machine learning and add a fifth stage that transforms the two pose estimates into a relative vector. Furthermore, we propose a novel supervised learning method using bounding box corrections such that our trained artificial neural networks can accurately predict 2D image points corresponding to known 3D object points. Simulation results show that this method is reliable, accurate (within 3 cm at contact), and fast (45.5 fps). [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a condition that degenerates over time and impairs speech and pronunciation because brain cells have died. This research work aims to predict parkinson disease using the voice features extracted from speech signals recorded from PD individuals with dysphonic speech disorders by employing deep learning algorithms. PD is challenging to diagnose early on in the clinical presentation. To address the issue in machine learning methods, this paper proposes a neural network model by processing speech signals to classify PD using the University of California Irvine (UCI) machine learning repository dataset. Initially, a pre-loss reduction module is created by using pre-sampling to make the dataset balanced by reducing the dimensionality and maintaining the size of the space without influencing the learning process for data preparation. The relevant features are derived using a novel multi-agent salp swarm (MASS) algorithm, and a novel Parkinson classification neural network (PCNN) is proposed to classify Parkinson's patients with high accuracy employing these derived features. The result shows that the models that use MASS-PCNN produce higher classification accuracy of 99.1%, precision of 97.8%, recall of 94.7% and F1-score of 0.995 when paralleled to the existing models. As an outcome, the suggested model will perform superior to common convolutional neural networks. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a neurological disease that has been reported to have affected most people worldwide. Recent research pointed out that about 90% of PD patients possess voice disorders. Motivated by this fact, many researchers proposed methods based on multiple types of speech data for PD prediction. However, these methods either face the problem of low rate of accuracy or lack generalization. To develop an approach that will be free of these issues, in this paper we propose a novel ensemble approach. These paper contributions are two folds. First, investigating feature selection integration with deep neural network (DNN) and validating its effectiveness by comparing its performance with conventional DNN and other similar integrated systems. Second, development of a novel ensemble model namely EOFSC (Ensemble model with Optimal Features and Sample Dependant Base Classifiers) that exploits the findings of recently published studies. Recent research pointed out that for different types of voice data, different optimal models are obtained which are sensitive to different types of samples and subsets of features. In this paper, we further consolidate the findings by utilizing the proposed integrated system and propose the development of EOFSC. For multiple types of vowel phonations, multiple base classifiers are obtained which are sensitive to different subsets of features. These features and sample-dependent base classifiers are integrated, and the proposed EOFSC model is constructed. To evaluate the final prediction of the EOFSC model, the majority voting methodology is adopted. Experimental results point out that feature selection integration with neural networks improves the performance of conventional neural networks. Additionally, feature selection integration with DNN outperforms feature selection integration with conventional machine learning models. Finally, the newly developed ensemble model is observed to improve PD detection accuracy by 6.5%. [ABSTRACT FROM AUTHOR]

Traditional computing architecture (Von Neumann) that requires data transfer between the off-chip memory and processor consumes a large amount of energy when running machine learning (ML) models. Memristive synaptic devices are employed to eliminate this inevitable inefficiency in energy while solving cognitive tasks. However, the performances of energy-efficient neuromorphic systems, which are expected to provide promising results, need to be enhanced in terms of accuracy and test error rates for classification applications. Improving accuracy in such ML models depends on the optimal learning parameter changes from a device to algorithm-level optimisation. To do this, this paper considers the Adadelta, an adaptive learning rate technique, to achieve accurate results by reducing the losses and compares the accuracy, test error rates, and energy consumption of stochastic gradient descent (SGD), Adagrad and Adadelta optimisation methods integrated into the Ag:a-Si synaptic device neural network model. The experimental results demonstrated that Adadelta enhanced the accuracy of the hardware-based neural network model by up to 4.32% when compared to the Adagrad method. The Adadelta method achieved the best accuracy rate of 94%, while DGD and SGD provided an accuracy rate of 68.11 and 75.37%, respectively. These results show that it is vital to select a proper optimisation method to enhance performance, particularly the accuracy and test error rates of the neuro-inspired nano-synaptic device-based neural network models. [ABSTRACT FROM AUTHOR]

ExSeisDat is designed using standard message passing interface (MPI) library for seismic data processing on high-performance super-computing clusters. These clusters are generally designed for efficient execution of complex tasks including large size IO. The IO performance degradation issues arise when multiple processes try accessing data from parallel networked storage. These complications are caused by restrictive protocols running by a parallel file system (PFS) controlling the disks and due to less advancement in storage hardware itself as well. This requires and leads to the tuning of specific configuration parameters to optimize the IO performance, commonly not considered by users focused on writing parallel application. Despite its consideration, the changes in configuration parameters are required from case to case. It adds up to further degradation in IO performance for a large SEG-Y format seismic data file scaling to petabytes. The SEG-Y IO and file sorting operations are the two of the main features of ExSeisDat. This research paper proposes technique to optimize these SEG-Y operations based on artificial neural networks (ANNs). The optimization involves auto-tuning of the related configuration parameters, using IO bandwidth prediction by the trained ANN models through machine learning (ML) process. Furthermore, we discuss the impact on prediction accuracy and statistical analysis of auto-tuning bandwidth results, by the variation in hidden layers nodes configuration of the ANNs. The results have shown the overall improvement in bandwidth performance up to 108.8% and 237.4% in the combined SEG-Y IO and file sorting operations test cases, respectively. Therefore, this paper has demonstrated the significant gain in SEG-Y seismic data bandwidth performance by auto-tuning the parameters settings on runtime by using an ML approach. [ABSTRACT FROM AUTHOR]

Skin cancer, primarily resulting from the abnormal growth of skin cells, is among the most common cancer types. In recent decades, the incidence of skin cancer cases worldwide has risen significantly (one in every three newly diagnosed cancer cases is a skin cancer). Such an increase can be attributed to changes in our social and lifestyle habits coupled with devastating man-made alterations to the global ecosystem. Despite such a notable increase, diagnosis of skin cancer is still challenging, which becomes critical as its early detection is crucial for increasing the overall survival rate. This calls for advancements of innovative computer-aided systems to assist medical experts with their decision making. In this context, there has been a recent surge of interest in machine learning (ML), in particular, deep neural networks (DNNs), to provide complementary assistance to expert physicians. While DNNs have a high processing capacity far beyond that of human experts, their outputs are deterministic, i.e., providing estimates without prediction confidence. Therefore, it is of paramount importance to develop DNNs with uncertainty-awareness to provide confidence in their predictions. Monte Carlo dropout (MCD) is vastly used for uncertainty quantification; however, MCD suffers from overconfidence and being miss calibrated. In this paper, we use MCD algorithm to develop an uncertainty-aware DNN that assigns high predictive entropy to erroneous predictions and enable the model to optimize the hyper-parameters during training, which leads to more accurate uncertainty quantification. We use two synthetic (two moons and blobs) and a real dataset (skin cancer) to validate our algorithm. Our experiments on these datasets prove effectiveness of our approach in quantifying reliable uncertainty. Our method achieved 85.65 ± 0.18 prediction accuracy, 83.03 ± 0.25 uncertainty accuracy, and 1.93 ± 0.3 expected calibration error outperforming vanilla MCD and MCD with loss enhanced based on predicted entropy. [ABSTRACT FROM AUTHOR]

Machine learning neural network (NN) algorithms are being applied for the past few years in all engineering and science domain, economic sectors, image processing synthesis and analysis, and so on. Due to this, this paper work considered employing these machine learning neural models for forecasting application in respect of renewable energy applications and in particular focused on the forecasting of wind speed. It is crucial for power grid dispatchability, stability, and controllability, and its precision is necessary for making the most use of wind resources. This article describes the development of a novel hybrid forecasting system to anticipate the wind speed of real-time wind farm datasets using a hybrid probabilistic neural network (PNN) model and optimization method. The particle swarm optimization (PSO)–ant lion optimization technique is utilized to modify the designed adaptive PNN in order to optimize the weight parameters. The machine learning model employed in this study is an adaptive PNN with a probability density function and a decision-making function that adhere to Bayes' rule to achieve faster convergence and higher prediction accuracy. The obtained simulation results show that the recommended hybrid optimized PNN model outperforms the other techniques that were evaluated and compared from the literature. This establishes that the built optimized adaptive PNN model is applicable and suitable to serve as a predictor, as shown by the outcome of the statistical analysis. [ABSTRACT FROM AUTHOR]

To reliably and accurately classify complicated "big" datasets, machine learning models must be continually improved. This research proposes straightforward yet competitive neural networks for text classification, even though graph neural networks (GNN) have reignited interest in graph-based text classification models. Convolutional neural networks (CNN), artificial neural networks (ANN), and their refined "fine-tuned" models (denoted as FT-CNN and FT-ANN) are the names given to our proposed models. The models presented in this paper demonstrate that our simple models like (CNN, ANN, FT-CNN, and FT-ANN) can perform better than more complex GNN ones such as (SGC, SSGC, and TextGCN) and are comparable to others (i.e., HyperGAT and Bert). The process of fine-tuning is also highly recommended because it improves the performance and reliability of models. The performance of our suggested models on five benchmark datasets (namely, Reuters (R8), R52, 20NewsGroup, Ohsumed, and Mr) is vividly illustrated. According to the experimental findings, on the majority of the target datasets, these models—especially those that have been fine-tuned—perform surprisingly better than SOTA approaches, including GNN-based models. [ABSTRACT FROM AUTHOR]

As a common data-driven method, artificial neural networks have been widely used in electricity spot price forecasting. To improve the accuracy of short-term forecasts, this paper proposes an optimized artificial neural network model for monthly electricity spot prices forecasting. A genetic algorithm is applied to regulate the weights and biases parameters of the artificial neural network structure. This study uses various historical dataset at monthly periods selected from Nordic electricity spot prices. For efficiency comparison, one-step ahead forecast method based on Schwartz-Smith stochastic model and two other prediction models, artificial neural network trained by Levenberg–Marquardt and particle swarm optimization algorithms are also presented. The comparison results show that the prediction model based on the genetic optimization algorithm is more accurate than the other prediction models. The proposed forecasting model can be considered as an alternative technique for the electricity spot price forecasting in the Nordic regions. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) are widely regarded as the most effective learning tool for dealing with large datasets, and they have been successfully used in thousands of applications in a variety of fields. Based on these large datasets, they are trained to learn the relationships between various variables. The adaptive moment estimation (Adam) algorithm, a highly efficient adaptive optimization algorithm, is widely used as a learning algorithm in various fields for training DNN models. However, it needs to improve its generalization performance, especially when training with large-scale datasets. Therefore, in this paper, we propose HN Adam, a modified version of the Adam Algorithm, to improve its accuracy and convergence speed. The HN_Adam algorithm is modified by automatically adjusting the step size of the parameter updates over the training epochs. This automatic adjustment is based on the norm value of the parameter update formula according to the gradient values obtained during the training epochs. Furthermore, a hybrid mechanism was created by combining the standard Adam algorithm and the AMSGrad algorithm. As a result of these changes, the HN_Adam algorithm, like the stochastic gradient descent (SGD) algorithm, has good generalization performance and achieves fast convergence like other adaptive algorithms. To test the proposed HN_Adam algorithm performance, it is evaluated to train a deep convolutional neural network (CNN) model that classifies images using two different standard datasets: MNIST and CIFAR-10. The algorithm results are compared to the basic Adam algorithm and the SGD algorithm, in addition to other five recent SGD adaptive algorithms. In most comparisons, the HN Adam algorithm outperforms the compared algorithms in terms of accuracy and convergence speed. AdaBelief is the most competitive of the compared algorithms. In terms of testing accuracy and convergence speed (represented by the consumed training time), the HN-Adam algorithm outperforms the AdaBelief algorithm by an improvement of 1.0% and 0.29% for the MNIST dataset, and 0.93% and 1.68% for the CIFAR-10 dataset, respectively. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a supervised learning algorithm, which avoids backward recursive gradient computation, for training deep convolutional spiking neural networks (SNNs) with single-spike-based temporal coding. The algorithm employs a linear approximation to compute the derivative of the spike latency with respect to the membrane potential, and it uses spiking neurons with piecewise linear postsynaptic potential to reduce the computational cost and the complexity of neural processing. To evaluate the performance of the proposed algorithm in deep architectures, we employ it in convolutional SNNs for the image classification task. For two popular benchmarks of MNIST and Fashion-MNIST datasets, the network reaches accuracies of, respectively, 99.2 and 92.8 % . The trade-off between memory storage capacity and computational cost with accuracy is analyzed by applying two sets of weights: real-valued weights that are updated in the backward pass and their signs, binary weights, that are employed in the feedforward process. We evaluate the binary CSNN on two datasets of MNIST and Fashion-MNIST and obtain acceptable performance with a negligible accuracy drop with respect to real-valued weights (about 0.6 and 0.8 % drops, respectively). [ABSTRACT FROM AUTHOR]

In this paper, conductive polymer-based composites in order to have higher electrical conductivity have been constructed using different nanoparticles and numerically considered by different classification techniques. Due to non-conducting feature of polymer-based composites, their other positive advantages (e.g., light weight and stress corrosion) underneath non-conducting defect in which this paper has tried to overcome the faced challenges. For this purpose, carbon black (CB), carbon nanotube (CNT), and expanded graphite (EG) with different weight percentages are added to the epoxy resin as input factors and the electrical conductivity of the samples are measured as response factor. The analysis of input factors is performed and the Taguchi method, artificial neural networks (ANNs) and extreme learning machine (ELM) are designed and used for the prediction of the response factor. The predicted responses using the applied methods are compared with the experimental results. In order to increase the mechanical strength, ten layers of unidirectional carbon fiber are used. The simulation results show that the ANNs and ELM provide good compatible predictions with respect to actual experiment data. Besides, obtained experimental results prove that the highest electrical conductivity has been achieved using 10, 15, and 25 percent using the CNT, EG, and CB, respectively. As a novelty of this paper, the constructed sample composite reaches the acceptable electrical conductivity suggested by United Stated Department of Energy standard considered as material development. In particular, the findings of this research can be used to construct conductive electrodes particularly in oil and gas industries. [ABSTRACT FROM AUTHOR]

English expression, language characteristics and usage norms are quite special, which is quite different from Chinese. This has special requirements for auxiliary teaching tools that use computer technology for English text processing. Based on neural network algorithm, this paper combines the actual needs of English grammar error correction to construct an English grammar error correction model based on neural network. In data processing, after feature selection, logistic regression model is used to analyze the influence of different features on article error correction. The article error correction incorporating word vector features mainly explores how to effectively express the features in English grammar error correction. In addition, this paper proposes two methods to optimize the feature representation in article error correction. One is to directly use the word vector corresponding to the word as a feature, replacing the original One-hot encoding, and the other uses a clustering method to compress the article features. Finally, this paper designs experiments to study the performance of the model constructed in this paper. The results obtained show that the model constructed in this paper has a certain effect. [ABSTRACT FROM AUTHOR]

How to build a football match prediction model and use scientific methods to solve the prediction problem has become a key point in the application of artificial intelligence in the sports industry. In this paper, we choose a BP neural network model that is powerful in processing nonlinear data to perform research. According to the demand, this paper constructs a gray fuzzy prediction model based on neural network, a gray extreme learning machine prediction model, and a gray fuzzy extreme learning machine prediction combination model based on neural network. Moreover, this paper tests the neural network model by comparing actual results with predicted results. In addition, by predicting and analyzing the football league data, this article tests the three models in terms of match result prediction accuracy, data processing speed, data transmission accuracy, match analysis scores, etc., and uses statistical analysis methods to process data, and uses intuitive statistical graphs to obtain the processing results. The research results show that the gray fuzzy extreme learning machine prediction combination model based on neural network constructed in this paper can retain the advantages of a single model and effectively improve the prediction accuracy of the model and the performance of the system. [ABSTRACT FROM AUTHOR]

In order to overcome the shortcomings of the current English network learning system, based on the neural network algorithm, this paper constructs an intelligent English network learning system based on the improved algorithm. Moreover, by analyzing the coupling between recurrent neural networks by contrast methods, this paper infers the coupling between recurrent neural networks. Moreover, this paper studies the continuous attractors of the autoencoder neural network and studies the continuous attractors of different types of autoencoder models. On this basis, this paper expands the existing model, adds the module of the interaction between the external input and the visible layer and studies the conditions required for the continuous attractor of the autoencoder model. In addition, on the basis of actual needs, this paper constructs the basic structure of the model and integrates it into the improved algorithm proposed in this paper to realize English online intelligent learning. Finally, this paper designs experiments to analyze the practical effects of this model and analyzes the experimental results through mathematical statistics. The research results show that the English network learning system constructed in this paper is effective. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) are mimicking computationally powerful biologically inspired models in which neurons communicate through sequences of spikes, regarded here as sparse binary sequences of zeros and ones. In neuroscience it is conjectured that time encoding, where the information is carried by the temporal position of spikes, is playing a crucial role at least in some parts of the brain where estimation of the spiking rate with a large latency cannot take place. Motivated by the efficiency of temporal coding, compared with the widely used rate coding, the goal of this paper is to develop and train an energy-efficient time-coded deep spiking neural network system. To ensure that the similarity among input stimuli is translated into a correlation of the spike sequences, we introduce correlative temporal encoding and extended correlative temporal encoding techniques to map analog input information into input spike patterns. Importantly, we propose an implementation where all multiplications in the system are replaced with at most a few additions. As a more efficient alternative to both rate-coded SNNs and artificial neural networks, such system represents a preferable solution for the implementation of neuromorphic hardware. We consider data classification tasks where input spike patterns are presented to a feed-forward architecture with leaky-integrate-and-fire neurons. The SNN is trained by backpropagation through time with the objective to match sequences of output spikes with those of specifically designed target spike patterns, each corresponding to exactly one class. During inference the target spike pattern with the smallest van Rossum distance from the output spike pattern determines the class. Extensive simulations indicate that the proposed system achieves a classification accuracy at par with that of state-of-the-art machine learning models. [ABSTRACT FROM AUTHOR]

The explosive growth of geospatial data is increasing requirements for automatic and efficient data learning abilities. Many deep learning methods have been widely applied for geospatial data understanding tasks, such as road networks and geospatial object detection. However, the demands for more accurate learning of high-level features require the use of deeper neural networks. To further improve the learning efficiency of deep neural networks, in this paper, we propose an improved convolutional neural network named EDense. First, we use its dense connectivity to integrate a CNN with an extreme learning machine. Then, we expand the kernels in the convolutional layers to increase the width of the network model. Furthermore, we propose one-feature EDense (OF-EDense), which is a simplified version of EDense, to fit conditions in which the number of parameters is strictly limited. Finally, the experimental results fully demonstrate the strong learning ability and high learning efficiency of EDense. [ABSTRACT FROM AUTHOR]

The application of Deep Neural Networks (DNNs) to a broad variety of tasks demands methods for coping with the complex and opaque nature of these architectures. When a gold standard is available, performance assessment treats the DNN as a black box and computes standard metrics based on the comparison of the predictions with the ground truth. A deeper understanding of performances requires going beyond such evaluation metrics to diagnose the model behavior and the prediction errors. This goal can be pursued in two complementary ways. On one side, model interpretation techniques "open the box" and assess the relationship between the input, the inner layers and the output, so as to identify the architecture modules most likely to cause the performance loss. On the other hand, black-box error diagnosis techniques study the correlation between the model response and some properties of the input not used for training, so as to identify the features of the inputs that make the model fail. Both approaches give hints on how to improve the architecture and/or the training process. This paper focuses on the application of DNNs to computer vision (CV) tasks and presents a survey of the tools that support the black-box performance diagnosis paradigm. It illustrates the features and gaps of the current proposals, discusses the relevant research directions and provides a brief overview of the diagnosis tools in sectors other than CV. [ABSTRACT FROM AUTHOR]

In machine learning models, one of the most popular models is artificial neural networks. The activation function is one of the important parameters of neural networks. In this paper, the sigmoid function is used as an activation function with a fractional derivative approach to minimize the convergence error in backpropagation and to maximize the generalization performance of neural networks. The proportional Caputo definition is considered a fractional derivative. We evaluated three neural network models on the usage of the proportional Caputo derivative. The results show that the proportional Caputo derivative approach has higher classification accuracy than traditional derivative models in backpropagation for neural networks with and without L2 regularization. [ABSTRACT FROM AUTHOR]

In this paper, we propose a method to approximate the solutions to nonlinear ordinary differential equations (ODE) using a deep learning feedforward artificial neural networks (ANNs). The efficiency of the proposed—unsupervised type machine learning—method is shown by solving two boundary value problems (BVPs) from quantum mechanics and nanofluid mechanics. The proposed mean-squared loss function is the sum of two terms: the first term satisfies the differential equation, while the second term satisfies the initial or boundary conditions. The total loss function is minimized by using general type of quasi-Newton optimization methods to get a desired network output. The approximation capability of the proposed method is verified for two sets of boundary value problems: first, a second-order nonlinear ODE and, second, a system of coupled nonlinear third-order ODEs. Point-wise comparison of our approximation shows a strong agreement with the available exact solutions and/or Runge–Kutta-based numerical solutions. We remark that the proposed algorithm minimizes the overall learnable network hyperparameters in a given initial or boundary value problems. More importantly, for the coupled system of third-order nonlinear ordinary differential equations, the proposed method does not need any adjustment with the initial/boundary conditions. Also, the current method does not require any special type of computational mesh. A straightforward minimization of total loss function yields a highly accurate results even with less number of epochs. Therefore, the proposed framework offers an attractive setting for the fluid mechanics community who are interested in studying heat and mass transfer problems. [ABSTRACT FROM AUTHOR]

The risk of death incurred by breast cancer is rising exponentially, especially among women. This made the early breast cancer detection a crucial problem. In this paper, we propose a computer-aided diagnosis (CAD) system, called CADNet157, for mammography breast cancer based on transfer learning and fine-tuning of well-known deep learning models. Firstly, we applied hand-crafted features-based learning model using four extractors (local binary pattern, gray-level co-occurrence matrix, and Gabor) with four selected machine learning classifiers (K-nearest neighbors, support vector machine, random forests, and artificial neural networks). Then, we performed some modifications on the Basic CNN model and fine-tuned three pre-trained deep learning models: VGGNet16, InceptionResNetV2, and ResNet152. Finally, we conducted a set of experiments using two benchmark datasets: Digital Database for Screening Mammography (DDSM) and INbreast. The results of the conducted experiments showed that for the hand-crafted features based CAD system, we achieved an area under the ROC curve (AUC) of 95.28% for DDSM using random forest and 98.10% for INbreast using support vector machine with the histogram of oriented gradients extractor. On the other hand, CADNet157 model (i.e., fine-tuned ResNet152) was the best performing deep model with an AUC of 98.90% (sensitivity: 97.72%, specificity: 100%), and 98.10% (sensitivity: 100%, specificity: 96.15%) for, respectively, DDSM and INbreast. The CADNet157 model overcomes the limitations of traditional CAD systems by providing an early detection of breast cancer and reducing the risk of false diagnosis. [ABSTRACT FROM AUTHOR]

Age estimation from face images can be profitably employed in several applications, ranging from digital signage to social robotics, from business intelligence to access control. Only in recent years, the advent of deep learning allowed for the design of extremely accurate methods based on convolutional neural networks (CNNs) that achieve a remarkable performance in various face analysis tasks. However, these networks are not always applicable in real scenarios, due to both time and resource constraints that the most accurate approaches often do not meet. Moreover, in case of age estimation, there is the lack of a large and reliably annotated dataset for training deep neural networks. Within this context, we propose in this paper an effective training procedure of CNNs for age estimation based on knowledge distillation, able to allow smaller and simpler "student" models to be trained to match the predictions of a larger "teacher" model. We experimentally show that such student models are able to almost reach the performance of the teacher, obtaining high accuracy over the LFW+, LAP 2016 and Adience datasets, but being up to 15 times faster. Furthermore, we evaluate the performance of the student models in the presence of image corruptions, and we demonstrate that some of them are even more resilient to these corruptions than the teacher model. [ABSTRACT FROM AUTHOR]

In the previous works on heat stress detection, various state-of-the-art machine learning techniques had been used to detect stress patterns from electroencephalographic (EEG) signals. Since the handcrafted feature engineering-based approaches pose certain limitations and are sensitive to transform the nonlinearities of EEG, deep learning techniques have drawn attention in the domain of EEG-based applications. Moreover, existing approaches for stress detection consider the whole frequency band (delta to gamma) which conceals the redundant and lossy information that increased the false detection rate. Also, these approaches were implemented only to identify the stress in binary classes and confined their application to identifying the level of stress. Therefore, a multi-scale channel-wise convolution-based multi-level heat stress assessment has been proposed in this paper. The novel contributions of this work are (1) designing a multi-scale convolutional neural network (MS-CNN) for extracting precise information from individual frequency bands of EEG, (2) use of sparse connectivity to reduce the redundant information and increase the performance with less trainable parameters, and (3) multi-level heat stress assessment for precise identification of the severity of stress. The proposed approaches were evaluated on pre-recorded data of 10 rats in a simulated laboratory environment. The high accuracy of approximately 96–97% and 90–92% has been achieved for binary and multi-level stress detection. There is approximately a 6 to 50% reduction in the trainable parameters with a 2% improvement of accuracy achieved on adopting channel-wise convolution over MS-CNN and deep convolutional neural network (DCNN). This demonstrates the effectiveness of the adopted multiscale feature extraction and sparse connectivity to improve the performance with a less complex model. [ABSTRACT FROM AUTHOR]

This paper introduces a developed method of an artificial neural networks (ANN) architecture for estimating the residual strength of clay soil. To implement this purpose, a database of input soil parameters is built, including liquid limit, plasticity index, A-line value, clay fraction, massive minerals, mica, kaolinite, and smectite, in which the output is the residual friction angle. The ANN model was developed by extensively analyzing a number of hidden layers and number of neurons in every layer, incorporating a statistical investigation of the model performance. The obtained results indicate that the ANN model is an outperformed and promising method based on various well-known indicators such as correlation coefficient, mean absolute error, and root mean square error. The achieved ANN model also gives higher estimation accuracy than those results in the literature. Finally, partial dependence plot 2-D was used for sensitivity analysis within the ANN algorithm to investigate the effect of coupled input variables on the estimated residual friction angle of the soil. It was found that A-line value, clay fraction, and massive minerals are the most important input parameters influencing the residual friction angle. [ABSTRACT FROM AUTHOR]

Crime forecasting has been one of the most complex challenges in law enforcement today, especially when an analysis tends to evaluate inferable and expanded crime rates, although a few methodologies for subsequent equivalents have been embraced before. In this work, we use a strategy for a time series model and machine testing systems for crime estimation. The paper centers on determining the quantity of crimes. Considering various experimental analyses, this investigation additionally features results obtained from a neural system that could be a significant alternative to machine learning and ordinary stochastic techniques. In this paper, we applied various techniques to forecast the number of possible crimes in the next 5 years. First, we used the existing machine learning techniques to predict the number of crimes. Second, we proposed two approaches, a modified autoregressive integrated moving average model and a modified artificial neural network model. The prime objective of this work is to compare the applicability of a univariate time series model against that of a variate time series model for crime forecasting. More than two million datasets are trained and tested. After rigorous experimental results and analysis are generated, the paper concludes that using a variate time series model yields better forecasting results than the predicted values from existing techniques. These results show that the proposed method outperforms existing methods. [ABSTRACT FROM AUTHOR]

The monitoring of physical activities and recognition of motion disorders belong to important diagnostical tools in neurology and rehabilitation. The goal of the present paper is in the contribution to this topic by (1) analysis of accelerometric signals recorded by wearable sensors located at specific body positions and by (2) implementation of deep learning methods to classify signal features. This paper uses the general methodology to analysis of accelerometric signals acquired during cycling at different routes followed by the global positioning system. The experimental dataset includes 850 observations that were recorded by a mobile device in the spine area (L3 verterbra) for cycling routes with the different slope. The proposed methodology includes the use of deep learning convolutional neural networks with five layers applied to signal values transformed into the frequency domain without specification of any signal features. The accuracy of discrimination between different motion patterns for the uphill and downhill cycling and recognition of 4 classes associated with different route slopes was 96.6% with the loss criterion of 0.275 for sigmoidal activation functions. These results were compared with those evaluated for selected sets of features estimated for each observation and classified by the support vector machine, Bayesian methods, and the two-layer neural network. The best cross-validation error of 0.361 was achieved for the two-layer neural network model with the sigmoidal and softmax transfer functions. Our methodology suggests that deep learning neural networks are efficient in the assessment of motion activities for automated data processing and have a wide range of applications, including rehabilitation, early diagnosis of neurological problems, and possible use in engineering as well. [ABSTRACT FROM AUTHOR]

Recognizing the plant disease and pests in its golden time is a highly critical problem to be addressed, since the herbalist can apply treatments within this period and save the agricultural product. In this paper, a deep learning approach to recognize the disease from the leaves of the plants has been pursued. A client-server system is proposed in which the server-side model can leverage huge deep CNN architectures to classify the diseases, whereas the client-side model is to be chosen among small deep CNN architectures with low number of parameters in order to be easily deployed on the end-user mobile devices with poor processing powers. Here, a novel knowledge distillation technique has been leveraged that improves the accuracy level of the small client-side model significantly. This technique distills the perception knowledge of a large model classifier and transfers this knowledge to the small model in order to perform a similar prediction capability. By applying this idea on Plantvillage dataset, we could achieve 97.58 % accuracy on a small MobileNet architecture which is very close to the accuracy of a large Xception model on the server with 99.73 % accuracy. Through applying this teacher-student idea, we could improve the classification rate of the state-of-the-art tiny model by 2.12 % . [ABSTRACT FROM AUTHOR]

Traffic accidents as a result of driver fatigue and drowsiness have caused many injuries and deaths. Therefore, driver fatigue detection and prediction system have been recognized as important potential research areas to prevent accidents caused by fatigue and drowsiness while driving. In this study, driver fatigue is determined by using behavior-based measurement information. Recent studies show that deep neural network is trending state-of-the-art machine learning approaches. Hence, we propose the deep belief network (DBN) model, a deep learning type, used for classification of the symptoms of fatigue in this study. DBN structure is a kind of neural network. The number of hidden layers within the network and the number of units in each hidden layer play important roles in the design of any neural network. Therefore, the hidden layer and the count of units in the DBN model designed in this paper have been selected as a result of various experiments. A greedy method has been adopted to adjust the structure of the deep belief network. Subsequently, the proposed DBN architecture test on KOU-DFD, YawDD and Nthu-DDD datasets. Comparative and experimental results concluded that the proposed DBN architecture is as robust as the other approaches found in the literature and achieves an accuracy rate of approximately 86%. [ABSTRACT FROM AUTHOR]

Automated continuous condition monitoring of industrial electrical machines to identify internal faults has become one of the critical research areas for the past decade. Among various defects, early-stage identification of insulation failure in stator winding is of notable demand as it often occurs and accounts for 37% of the overall motor failures. Identifying the current signature at its embryonic stage will effectively improve industrial machinery's downtime and repair costs. Recent advances in computational performance and sensor technology concede advanced systems for achieving these goals. The design of an AI-based fault signature analyzer (FSA) has been developed in this paper. FSA uses real-time stator current data in the time and frequency domain from healthy and faulty induction motors to train the various AI-based machine learning classifiers to identify health conditions using wavelets. Comparing machine learning algorithms such as artificial neural network, random forest, fuzzy logic, neuro-fuzzy logic, K-nearest neighbors is performed, and various performance attributes are quantified. A reliable, automatic fault signature from a motor current is thus analyzed using the fusion of a wavelet-based feature extraction technique and a capable knowledge-based efficient artificial intelligence (AI) approach. [ABSTRACT FROM AUTHOR]

Solutions in geotechnics have been optimizing with the aid of machine learning methods. The aim of this paper is to apply different machine learning algorithms in order to achieve rock mass classification. It is demonstrated that RMR classification system can be obtained using only variables which are closely related to rock mass quality, instead of all RMR variables, without missing significant accuracy. The different machine learning algorithms used are the naïve Bayes, random forest, artificial neural networks and support vector machines. The variables to calculate RMR, selected by factor analysis, are: rock strength, rock weathering, spacing, persistence and aperture of discontinuities and presence of water. The machine learning models were trained and tested thirty times, with random subsampling, using two-thirds of the total database for training sample. The models presented average accuracy greater than 0.81, which was calculated from the confusion matrix, using the proportion of true positives and true negatives in the test sample. Significant values of efficiency, precision and reproducibility rates were achieved. The study shows the application of machine learning algorithms allows obtaining the RMR classes, even with a small number of variables. In addition, the results of the evaluation metrics of the developed algorithms show that the methodology can be applied to new database, working as a valuable way to achieve rock mass classification. [ABSTRACT FROM AUTHOR]

Active Learning Method (ALM) is one of the powerful tools in soft computing and it is inspired by the human brain capabilities in approaching complicated problems. ALM, which is in essence an adaptive fuzzy learning algorithm, tries to model a Multi-Input Single-Output system with several single-input single-output subsystems. Each of these subsystems is then modeled by an ink drop spread (IDS) plane. IDS operator, which is the main processing engine of ALM, extracts two kinds of informative features, Narrow Path and Spread, from each IDS plane without complicated computations. These features from all IDS planes are then aggregated in the inference engine. Despite the great performance of ALM in different applications, an efficient hardware implementation has remained a challenge, which is mainly due to considerably high memory requirement of IDS operation. In this paper, in a novel approach to IDS operation, we propose an abstract representation of the IDS planes which minimizes the memory requirement and the computational cost, and consequently, benefits the hardware implementation in terms of area and speed. The proposed approach is fully compatible with memristor-crossbar implementation with an adaptive learning capability. Simpler learning algorithm and higher speed make our proposed algorithm suitable for applications where real-time process, low-cost and small implementation are of high priority. Applications in the classification of real-world datasets and function approximation are provided to confirm the effectiveness of the algorithm. Eventually, the paper concludes that the proposed computing structure provides a synergy between artificial neural networks and fuzzy domains. [ABSTRACT FROM AUTHOR]

In this paper, a novel combination of deep learning recurrent neural network and Lyapunov time is proposed to forecast the consumption of electricity load, in Greece, in normal/abrupt change value areas. Our method verifies the chaotic behavior of load time series through chaos time series analysis and with the application of deep learning recurrent neural networks produces predictions for 10 and 20 days ahead. Specifically, four different neural network models constructed (a) feed forward neural network, (b) gated recurrent unit (GRU) neural network, (c) long short-term memory (LSTM) recurrent and (d) bidirectional LSTM neural network to implement the prediction in a prediction horizon, produced through the extraction of maximum Lyapunov exponent. We constructed sequences of algorithms to feed the neural networks, creating three scenarios (a) 1-step, (b) 10-step and (c) 20-step sequences. For each neural network model, we used its predictions as inputs to predict steps forward, iteratively, to examine the accuracy of the proposed models, for horizons that are both inside and outside to that defined by Lyapunov time. The results show that the deep learning GRU neural network produces iterative predictions of high accuracy and stability, following the trend evolution of actual values, even outside the safe horizon for 1-step and 10-step cases. [ABSTRACT FROM AUTHOR]

Predicting the critical buckling loads of functionally graded material (FGM) plates using an analytical method requires solving complex equations with various modes of deformation to determine the minimum loads. The approach is too complex for application in engineering practice. In this paper, a data-driven model using the artificial neural network (ANN) is proposed for the critical buckling load of FGM plates, as an alternative tool for practicing engineers. A database is first developed for randomly selected inputs using an analytical solution based on first-order shear deformation theory for simply supported FGM plates. The database is then divided into a training dataset with 80% of the data and a testing dataset with 20% of the data for developing and validating, respectively, the ANN model. The ANN model developed using six hidden layers with 32 nodes in each layer is found to match the data with a coefficient of determination of 99.95%. Using the ANN model, the stochastic characteristic of the critical buckling load is examined with respect to randomness of the input parameters. The study reveals that along with the dimensional parameters, the critical buckling load is significantly affected by the randomness of the volume fraction ratio and ratio of the modulus of elasticity of the ceramic and the metal. [ABSTRACT FROM AUTHOR]

Sand–cement stabilization is considered as one of the most common in situ methods in the soil improvement practices. Despite the importance of studying some fundamental characteristics such as the stress (q)–strain (ε ) and the pore pressure (u)–strain (ε ) behavior of the stabilized soils, very few studies have examined this so far. Hence, this paper aims at adopting an initiative approach which is group method of data handling (GMDH)-type neural network to specifically predict such behavior for cement-treated sands using the consolidated undrained (CU) triaxial test results. To do so, the q– ε and u– ε results from the CU tests are considered on the basis of different variables such as cement content (C), confining pressure (CP), porosity (η ) and curing time (D). The obtained data, regarding similar statistical characteristics, are randomly sorted into three groups namely training, validation and testing. Current modeling is based on the first group (80% of the data), whereas the comparisons are made among other approaches in terms of the last one. Moreover, to achieve more accurate predictions, parameters related to the stress ( q n - 1 ) and pore pressure ( u n - 1 ) in the previous strain level are assumed in the modeling. It can be concluded that the two-hidden layer model is capable of accurately predicting the q– ε and u– ε behavior for the testing data, compared to other machine learning methods. By and large, GMDH modeling is strongly suggested as a potent method to estimate the soil mechanical properties like brittle index (IB), maximum strength (qmax), failure strain (ε f) and stiffness (E50). [ABSTRACT FROM AUTHOR]

The ecological footprint is an excellent tool to better understand the consequences of the human behavior on the environment. The growing need for natural resources emphasizes the necessity of their accurate observation, calculation, and prediction. This paper develops and compares four hybrid machine learning models for predicting the total ecological footprint of consumption based on a set of hyper-parameters predefined by the Bayesian optimization algorithm. In particular, K-nearest neighbor regression (KNNReg), random forest regression (RFR) with 93 trees, and two artificial neural networks (ANNs) with two hidden layers were developed and later compared in terms of their performance. As energy inputs, the primary energy consumption from (1) natural gas sources, (2) coal sources, (3) oil sources, (4) wind sources, (5) solar photovoltaic sources, (6) hydropower sources, (7) nuclear sources, and (8) other renewable sources was used. Additionally, population number has also been used as an input. The models were developed using a set of data that include 1804 instances. The ANNs were modeled using two different activation functions in the hidden layers: ReLU and SPOCU. The performance was evaluated using the mean absolute percentage error (MAPE), mean absolute scaled error (MASE), normalized root-mean-squared error (NRMSE), and symmetric mean absolute percentage error (SMAPE). The results show that KNNReg performs the best with MASE of 0.029, followed by the RFR (0.032), ReLU ANN (0.064), and SPOCU ANN (0.089). Moreover, SMOGN was utilized to produce a synthetic test set which was used to additionally test the best performed model. The performance on the SMOGN set demonstrates good performance (MASE=0.022). Lastly, the best performed model was implemented into a GUI that calculates the ecological footprint based on user inputs. [ABSTRACT FROM AUTHOR]

The analysis on economic data based on time series takes an important position in the field of analysis on time-series data and is also an important task of the field of big data and artificial intelligence. Traditional time-series analysis method is of relatively weak competence in dealing with multi-series analysis. In this research, based on the problem associated with the analysis on time-series economic data, efficient handling method and model are put forward in the face of multi-series analysis task. Also, combined with the association rules, trend correlation and self-trend correlation among multiple series, a trend and correlation deep neural network model (TC-DNM) is established and then tested and verified by using three kinds of economic datasets with representativeness based on the trend analysis task handed by multi-series analysis. The results show that the model proposed in this research is effective than a number of baseline models, can be employed to achieve precision–recall balance and also possesses strong reusability. The two correlation models and joint models in this paper are of peculiarity and innovativeness. [ABSTRACT FROM AUTHOR]

With the development of economic science and technology, the development of computer vision has undergone rapid changes, and various products relying on computer vision are also more and more, such as smart home, robot technology, and so on. At present, robot technology has become a very important part of the development of human science and technology, and in the field of industrial robots, the most rapid development is the robot with robot arm adaptive motion. It is very necessary to study the adaptive motion control of the manipulator based on machine learning. The robot with the adaptive motion of the manipulator can carry out logistics express sorting, operate in the doors and windows outside the building, and pick fruits in the orchard, which can ensure the effective implementation of hard work. Therefore, this paper proposes a mechanical adaptive control method based on a neural network. According to the motion model of the manipulator, the RBF neural network model is used to judge the stability of the system according to the Lyapunov function. The related algorithms of machine learning and multi-degree of freedom manipulator are studied and improved. The RBF neural network model approximates the unknown function infinitely and then establishes the complex motion model. Aiming at the adaptive neural network of a manipulator, a network adaptive terminal control method is proposed. Firstly, a stable manipulator motion system is designed by using a neural network, and then the terminal synovial controller is designed by using backstepping control technology. The stability of the method is proved by using the approximation virtual control technology of the neural network. The adaptive control is realized by using the learning and self-adaptability of the neural network; thus, the stability analysis of the closed-loop system is realized. [ABSTRACT FROM AUTHOR]

In order to asymptotically track the desired trajectory of nonlinear systems with unidirectional control input under various uncertainties, a modified saturated learning-based control law with adaptive notch filter (ANF) is presented in this paper. The proposed control law ensures the asymptotic convergence of the tracking error by utilizing a saturated function and a repetitive learning-based estimator, which allows the learning of the periodic dynamic behavior of the nonlinear system online to make compensations. Simultaneously, an ANF is integrated with the proposed control law to estimate the frequency of the periodic dynamics online. Differing from the existing methodologies in the literatures, the control law proposed in this paper only requires the unknown dynamics be bounded (the parameters, structure, and period of the unknown dynamics is not required to be known). Finally, applying the proposed control law to the electromagnetic suspension system, simulation, and experimental results are included to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Recent years have witnessed the exponential growth of time series data as the popularity of sensing devices and development of IoT techniques; time series classification has been considered as one of the most challenging studies in time series data mining, attracting great interest over the last two decades. According to the empirical evidences, temporal representation learning-based time series classification has more superiority of accuracy, efficiency and interpretability as compared to hundreds of existing time series classification methods. However, due to the high time complexity of feature process, the performance of these methods has been severely restricted. In this paper, we first presented an efficient shapelet transformation method to improve the overall efficiency of time series classification, and then, we further developed a novel enhanced recurrent neural network model for deep representation learning to further improve the classification accuracy. Experimental results on typical real-world datasets have justified the superiority of our models over several shallow and deep representation learning competitors. [ABSTRACT FROM AUTHOR]

Brain–computer interface (BCI) is the current trend in technology expansion as it provides an easy interface between human brain and machine. The demand for BCI-based applications is growing tremendously, and efforts are in progress to deploy BCI devices for real-world applications. One of the widely known applications of BCI technology is rehabilitation in which BCI devices can provide various types of assistance to specially abled persons. In this paper, the effect of hand actions on objects is analyzed for motor-related mental task. The proposed approach analyzes electroencephalogram (EEG)-based brain activity which was captured for images shown with different gripping actions on objects. The EEG recordings are first pre-processed, followed by extraction of epochs and frequency bands using discrete wavelet transform; afterward, feature extraction followed by training and classification steps is performed for classifying the grip action into congruent (correct) and incongruent (incorrect) grip categories. The proposed work makes use of average power and relative wavelet energy as discriminating features which are then fed to train an artificial neural network for automatically classifying the incoming EEG patterns into correct or incorrect object hand grips. The performance evaluation of the proposed system is done on real EEG dataset obtained from 14 subjects. Experimental results have shown an accuracy of 75%. Also, to evaluate the effectiveness of our work, a comparison of our work with other state-of-the-art works reported by different authors is presented at the end. The results show the effectiveness of proposed approach and suggest further that the system can be used to analyze and train subjects having motor-related disabilities for perceiving correct or incorrect hand grips on objects. [ABSTRACT FROM AUTHOR]

Misclassification is a critical problem in many machine learning applications. Since even the classifier models with high accuracy (e.g., > 95%) still introduce some misclassification error, it may not be possible to rely on the output of a classifier. In this paper, we introduce trustable learning, which prompts the learning model to yield only the true output, thus avoiding misclassifications. Whenever the model cannot decide the output accurately, the learning model should indicate that there could be a misclassification error if it is forced to classify, and hence, it should reject to make a decision or defer it to a human expert. Therefore, we develop a methodology for trustable learning and apply it to artificial neural networks and show that it is possible to develop a classifier with 0% misclassification error. We propose a novel neural network architecture named WisdomNet that could provide zero prediction error by introducing an additional neuron named as conjugate neuron that would indicate whether the network is able to classify the data correctly or not. The WisdomNet architecture can be applied to any previously built model, and we have evaluated WisdomNet with several network architectures such as multilayer perceptron, convolutional neural network, and deep network on different data sets. The results show that the WisdomNet is able to reduce the classification error rate to 0%, while labeling the data is difficult to classify as 'reject' at a low percentage of within around 10%. [ABSTRACT FROM AUTHOR]

Artificial neural networks (ANNs) have emerged as hot topics in the research community. Despite the success of ANNs, it is challenging to train and deploy modern ANNs on commodity hardware due to the ever-increasing model size and the unprecedented growth in the data volumes. Particularly for microarray data, the very high dimensionality and the small number of samples make it difficult for machine learning techniques to handle. Furthermore, specialized hardware such as graphics processing unit (GPU) is expensive. Sparse neural networks are the leading approaches to address these challenges. However, off-the-shelf sparsity-inducing techniques either operate from a pretrained model or enforce the sparse structure via binary masks. The training efficiency of sparse neural networks cannot be obtained practically. In this paper, we introduce a technique allowing us to train truly sparse neural networks with fixed parameter count throughout training. Our experimental results demonstrate that our method can be applied directly to handle high-dimensional data, while achieving higher accuracy than the traditional two-phase approaches. Moreover, we have been able to create truly sparse multilayer perceptron models with over one million neurons and to train them on a typical laptop without GPU (https://github.com/dcmocanu/sparse-evolutionary-artificial-neural-networks/tree/master/SET-MLP-Sparse-Python-Data-Structures), this being way beyond what is possible with any state-of-the-art technique. [ABSTRACT FROM AUTHOR]

This paper proposes a multilingual audio information management system based on semantic knowledge in complex environments. The complex environment is defined by the limited resources (financial, material, human, and audio resources); the poor quality of the audio signal taken from an internet radio channel; the multilingual context (Spanish, French, and Basque that is in under-resourced situation in some areas); and the regular appearance of cross-lingual elements between the three languages. In addition to this, the system is also constrained by the requirements of the local multilingual industrial sector. We present the first evolutionary system based on a scalable architecture that is able to fulfill these specifications with automatic adaptation based on automatic semantic speech recognition, folksonomies, automatic configuration selection, machine learning, neural computing methodologies, and collaborative networks. As a result, it can be said that the initial goals have been accomplished and the usability of the final application has been tested successfully, even with non-experienced users. [ABSTRACT FROM AUTHOR]

Stock price prediction is a rich research topic that has attracted interest from various areas of science. The recent success of machine learning in speech and image recognition has prompted researchers to apply these methods to asset price prediction. The majority of literature has been devoted to predicting either the actual asset price or the direction of price movement. In this paper, we study a hitherto little explored question of predicting significant changes in stock price based on previous changes using machine learning algorithms. We are particularly interested in the performance of neural network classifiers in the given context. To this end, we construct and test three neural network models including multilayer perceptron, convolutional net, and long short-term memory net. As benchmark models, we use random forest and relative strength index methods. The models are tested using 10-year daily stock price data of four major US public companies. Test results show that predicting significant changes in stock price can be accomplished with a high degree of accuracy. In particular, we obtain substantially better results than similar studies that forecast the direction of price change. [ABSTRACT FROM AUTHOR]

This paper addresses the analog optimization for non-differential functions. The Lagrange programming neural network (LPNN) approach provides us a systematic way to build analog neural networks for handling constrained optimization problems. However, its drawback is that it cannot handle non-differentiable functions. In compressive sampling, one of the optimization problems is least absolute shrinkage and selection operator (LASSO), where the constraint is non-differentiable. This paper considers the hidden state concept from the local competition algorithm to formulate an analog model for the LASSO problem. Hence, the non-differentiable limitation of LPNN can be overcome. Under some conditions, at equilibrium, the network leads to the optimal solution of the LASSO. Also, we prove that these equilibrium points are stable. Simulation study illustrates that the proposed analog model and the traditional digital method have the similar mean squared performance. [ABSTRACT FROM AUTHOR]

Artificial metaplasticity is the machine learning algorithm inspired in the biological metaplasticity of neural synapses. Metaplasticity stands for plasticity of plasticity, and as long as plasticity is related to memory, metaplasticity is related to learning. Implemented in supervised learning assuming input patterns distribution or a related function, it has proved to be very efficient in performance and in training convergence for multidisciplinary applications. Now, for the first time, this kind of artificial metaplasticity is implemented in an unsupervised neural network, achieving also excellent results that are presented in this paper. To compare results, a modified self-organization map is applied to the classification of MIT-BIH cardiac arrhythmias database. [ABSTRACT FROM AUTHOR]

In this paper, the concept of Virtual Weather Stations (VWS) is introduced. A VWS is an integration of algorithms to download meteorological data, process and use them with the main objective of estimate data in nearby locations with no meteorological stations. To develop the VWS, the performances of different interpolation methods were evaluated to test the accuracy. Daily data from an automatic weather station network, such as precipitation (Precip), air temperature (Temp), air relative humidity, mean wind speed, total solar irradiation, and reference evapotranspiration were interpolated using artificial neural networks (ANNs) with the hardlim, sigmoid, hyperbolic tangent (tanh), softsign, and rectified linear unit (relu) activations functions were employed. To contrast the ANNs interpolations, alternatives methods such as inverse distance weighting, inverse-squared distance weighting, multilinear regression, and random forest regression were used. To validate the models, a randomly selected weather station was removed from the daily datasets, and the interpolated values were compared with the actual station records. Additionally, interpolations in the summer and winter months were performed to check the capability of the models during periods with more extreme phenomena. The results showed that the interpolation methods have an R2 up to 0.98 for variables such as temperatures for the period of 1 year. Meanwhile, during the summer and winter, the models presented lower accuracy. From a practical perspective, the methods here described could be useful to produce meteorological data with the VWS to record temperatures and dose the irrigation in crops. [ABSTRACT FROM AUTHOR]

Hurricanes are among the most destructive natural phenomena on Earth. Timely prediction and tracking of hurricane intensity is important as it can help authorities in emergency planning. Several manual, semi and fully automated techniques based on different principles have been developed for hurricane intensity estimation. In this paper, a deep convolutional neural network architecture is proposed for fully automated hurricane intensity estimation from satellite infrared (IR) images. The proposed architecture is robust to errors in annotation of the storm center with a smaller root-mean-squared error (RMSE) (8.82 knots) in comparison to the previous state of the art methods. A web server implementation of Deep-PHURIE and its pre-trained neural network model are available at the URL: http://faculty.pieas.edu.pk/fayyaz/software.html#Deep-PHURIE. [ABSTRACT FROM AUTHOR]

This paper evaluates the potential of five modeling approaches, namely M5 model tree, random forest, artificial neural networks, support vector machines and Gaussian processes, for the prediction of unconfined compressive strength of stabilized pond ashes with lime and lime sludge. The study not only presents five models for the same set of data but also compares the overall performance of them. Dataset used consists of 255 samples acquired from laboratory experiments. Out of the total, 170 randomly chosen samples were used for training and remaining 85 were used for testing the models. Input dataset consists of eight parameters (uniformity coefficient, coefficient of curvature, maximum dry density, optimum moisture content, lime, lime sludge, curing period and 7-day soaked California bearing ratio), while the output is UCS value at 7, 28, 45, 90 and 180 days of curing. Comparisons of results propose that Gaussian processes modeling strategy works well and the overall performance was substantially nearer to the exact agreement line. As a result of GP model, higher value of CC = 0.997 and lower values of RMSE = 23.016 kPa and MAE = 16.455 were obtained for testing the dataset. Sensitivity analysis suggests that lime, lime sludge, curing period and California bearing ratio are the significant parameters for predicting the unconfined compressive strength of stabilized pond ashes. The results confirmed that GP models are in a position to predict the unconfined compressive strength of stabilized pond ashes with an excessive degree of accuracy; however, GP modeling approach proves that this approach is more economical and less difficult in comparison with tedious laboratory work. [ABSTRACT FROM AUTHOR]