One of the most effective image processing techniques is the use of convolutional neural networks, where we combine intensity values at grid points in the vicinity of each point. To speed up computations, researchers have developed a dilated version of this technique, in which only some points are processed. It turns out that the most efficient case is when we select points from a sub-grid. In this paper, we explain this empirical efficiency proving that the sub-grid is indeed optimal – in some reasonable sense. To be more precise, we prove that all reasonable optimality criteria, the optimal subset of the original grid is either a sub-grid, or a sub-grid-like set.

Optical artificial neural networks (ONNs) have significant potential for ultra-high computing speed and energy efficiency. We report a new approach to ONNs based on integrated Kerr micro-combs that is programmable, highly scalable and capable of reaching ultra-high speeds, demonstrating the building block of the ONN — a single neuron perceptron — by mapping synapses onto 49 wavelengths to achieve a single-unit throughput of 11.9 Giga-OPS at 8 bits per OP, or 95.2 Gbps. We test the perceptron on handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. By scaling the perceptron to a deep learning network using off-the-shelf telecom technology we can achieve high throughput operation for matrix multiplication for real-time massive data processing.

Multiply-accumulate calculations using a memristor crossbar array is an important method to realize neuromorphic computing. However, the memristor array fabrication technology is still immature, and it is difficult to fabricate large-scale arrays with high-yield, which restricts the development of memristor-based neuromorphic computing technology. Therefore, cascading small-scale arrays to achieve the neuromorphic computational ability that can be achieved by large-scale arrays, which is of great significance for promoting the application of memristor-based neuromorphic computing. To address this issue, we present a memristor-based cascaded framework with some basic computation units, several neural network processing units can be cascaded by this means to improve the processing capability of the dataset. Besides, we introduce a split method to reduce pressure of input terminal. Compared with VGGNet and GoogLeNet, the proposed cascaded framework can achieve 93.54% Fashion-MNIST accuracy under the 4.15M parameters. Extensive experiments with Ti/AlOx/TaOx/Pt we fabricated are conducted to show that the circuit simulation results can still provide a high recognition accuracy, and the recognition accuracy loss after circuit simulation can be controlled at around 0.26%.

Numerous approaches address over-fitting in neural networks: by imposing a penalty on the parameters of the network (L1, L2, etc); by changing the network stochastically (drop-out, Gaussian noise, etc.); or by transforming the input data (batch normalization, etc.). In contrast, we aim to ensure that a minimum amount of supporting evidence is present when fitting the model parameters to the training data. This, at the single neuron level, is equivalent to ensuring that both sides of the separating hyperplane (for a standard artificial neuron) have a minimum number of data points — noting that these points need not belong to the same class for the inner layers. We firstly benchmark the results of this approach on the standard Fashion-MINST dataset, comparing it to various regularization techniques. Interestingly, we note that by nudging each neuron to divide, at least in part, its input data, the resulting networks make use of each neuron, avoiding a hyperplane completely on one side of its input data (which is equivalent to a constant into the next layers). To illustrate this point, we study the prevalence of saturated nodes throughout training, showing that neurons are activated more frequently and earlier in training when using this regularization approach. A direct consequence of the improved neuron activation is that deep networks are now easier to train. This is crucially important when the network topology is not known a priori and fitting often remains stuck in a suboptimal local minima. We demonstrate this property by training a network of increasing depth (and constant width): most regularization approaches will result in increasingly frequent training failures (over different random seeds) whilst the proposed evidence-based regularization significantly outperforms in its ability to train deep networks.

Optical artificial neural networks (ONNs) — analog computing hardware tailored for machine learning [1,2] — have significant potential for ultra-high computing speed and energy efficiency [3]. We propose a new approach to architectures for ONNs based on integrated Kerr micro-comb sources [4] that is programmable, highly scalable and capable of reaching ultra-high speeds. We experimentally demonstrate the building block of the ONN — a single neuron perceptron — by mapping synapses onto 49 wavelengths of a micro-comb to achieve a high single-unit throughput of 11.9 Giga-FLOPS at 8 bits per FLOP, corresponding to 95.2 Gbps. We test the perceptron on simple standard benchmark datasets — handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. This performance is a direct result of the record small wavelength spacing (49GHz) for a coherent integrated microcomb source, which results in an unprecedented number of wavelengths for neuromorphic optics. Finally, we propose an approach to scaling the perceptron to a deep learning network using the same single micro-comb device and standard off-the-shelf telecommunications technology, for high-throughput operation involving full matrix multiplication for applications such as real-time massive data processing for unmanned vehicle and aircraft tracking.

Creating self-learning algorithms, developing deep neural networks and improving other methods that "learn" for various areas of human activity is the main goal of the theory of machine learning. It helps to replace the human with a machine, aiming to increase the quality of production. The theory of artificial neural networks, which already have replaced the humans in problems of detection of moving objects, recognition of images or sounds, time series prediction, big data analysis and numerical methods remains the most dispersed branch of the theory of machine learning. Certainly, for each area of human activity it is necessary to select appropriate neural network architectures, methods of data processing and some novel tools from applied mathematics. But the universal problem for all these neural networks with specific data is the achieving the highest accuracy in short time. Such problem can be resolved by increasing sizes of architectures and improving data preprocessing, where the accuracy rises with the training time. But there is a possibility to increase the accuracy without time growing, applying optimization methods. In this survey we demonstrate existing optimization algorithms of all types, which can be used in neural networks. There are presented modifications of basic optimization algorithms, such as stochastic gradient descent, adaptive moment estimation, Newton and quasi-Newton optimization methods. But the most recent optimization algorithms are related to information geometry, for Fisher-Rao and Bregman metrics. This approach in optimization extended the theory of classic neural networks to quantum and complex-valued neural networks, due to geometric and probabilistic tools. There are provided applications of all introduced optimization algorithms, what delighted many kinds of neural networks, which can be improved by including any advanced approaches in minimization of the loss function. Afterwards, we demonstrated ways of developing optimization algorithms in further researches, engaging neural networks with progressive architectures. Classical gradient based optimizers can be replaced by fractional order, bilevel and, even, gradient free optimization methods. There is a possibility to add such analogues in graph, spiking, complex-valued, quantum and wavelet neural networks. Besides the usual problems of image recognition, time series prediction, object detection, there are many are other tasks for modern theory of machine learning, such as solving problem of quantum computations, partial differential and integro-differential equations, stochastic processes and Brownian motion, making decisions and computer algebra.

This paper investigates the dynamic properties of artificial neural networks using differential equations and explores the influence of parameters on stability and neural oscillations. By analyzing the equilibrium point of the differential equations, we identify conditions for asymptotic stability and criteria for oscillation in artificial neural networks. Furthermore, we demonstrate how adjusting synaptic weights between neurons can effectively control stability and oscillation. The proposed model offers potential insights into the malfunctioning mechanisms of biological neural networks implicated in neurological disorders like Parkinson's disease tremors and epilepsy seizures, which are characterized by abnormal oscillations.

Multilabel learning goes beyond standard supervised learning models by associating a sample with more than one class label. Among the many techniques developed in the last decade to handle multilabel learning best approaches are those harnessing the power of ensembles and deep learners. This work proposes merging both methods by combining a set of gated recurrent units, temporal convolutional neural networks, and long short-term memory networks trained with variants of the Adam optimization approach. We examine many Adam variants, each fundamentally based on the difference between present and past gradients, with step size adjusted for each parameter. We also combine Incorporating Multiple Clustering Centers and a bootstrap-aggregated decision trees ensemble, which is shown to further boost classification performance. In addition, we provide an ablation study for assessing the performance improvement that each module of our ensemble produces. Multiple experiments on a large set of datasets representing a wide variety of multilabel tasks demonstrate the robustness of our best ensemble, which is shown to outperform the state-of-the-art. The MATLAB code for generating the best ensembles in the experimental section will be made available at https://github.com/LorisNanni.

Image denoising has been one of the important problems in the field of computer vision, and it has a wide range of practical value in many applications, such as medical image processing, image enhancement, and computational photography. Traditional image denoising methods are usually based on hand-designed features and filters, but these methods perform poorly under complex noise and image structures. In recent years, the rapid development of neural network technology has revolutionized the image-denoising task. This paper introduces the knowledge about neural networks and image denoising, explores the impact of neural networks on image denoising, and how is it possible to denoise images by neural networks. It also summarises other image-denoising methods and finally points out the challenges and problems faced by image-denoising at present. Some possible new development directions are proposed to provide new solutions for image-denoising researchers and to promote the development of the field.

Dynamic neural networks (DNN) are types of artificial neural networks (ANN) that are designed to work with sequential data where context in time is important. In contrast to traditional static neural networks that process data in a fixed order, dynamic neural networks use information about past inputs, which is important if dynamic of a certain process is emphasized. They are widely used in natural language processing, speech recognition and time series prediction. In industrial processes, their use is interesting for the prediction of difficult-to-measure process variables. In an industrial process of isomerization, it is crucial to measure the quality attributes affecting the octane number of gasoline. Process analyzers that are commonly used for this purpose are expensive and subject to failures, therefore, in order to achieve continuous production in case of malfunction, mathematical models for estimating the product quality attributes are imposed as a solution. In this paper, mathematical models were developed using dynamic recurrent neural networks (RNN), i.e., their subtype of a long short-term memory (LSTM) architecture. The results of the developed models were compared with the results of several types of other data-driven models developed for an isomerization process, such as multilayer perceptron (MLP) artificial neural networks, support vector machines (SVM) and dynamic polynomial models. The obtained results are satisfactory, which suggests a good possibility of application.

This study aimed at exploring what role artificial intelligence techniques could play in the futural numerical analysis. In this paper, a convolutional neural network techniques based on modified loss function is proposed as a surrogate of finite element method(FEM). Several surrogate-based physics-informed neural networks(PINNs) are developed to solve representative boundary value problems based on elliptic partial differential equations (PDEs). Results from the proposed surrogate-based approach are in good agreement with ones from conventional FEM. It is found that modification of the loss function could improve the prediction accuracy of the neural network. It is indicated that to some extent the artificial intelligence technique could replace conventional numerical analysis as a great surrogate model.

In computer vision and image analysis, Convolutional Neural Networks (CNNs) and other deep learning models are at the forefront of research and development. These advanced models have proven to be highly effective in tasks related to computer vision. One technique that has gained prominence in recent years is the construction of ensembles using Deep CNNs. These ensembles typically involve combining multiple pre-trained CNNs to create a more powerful and robust network. The purpose of this study is to evaluate the effectiveness of building CNN ensembles by combining several advanced techniques. Tested here are CNN ensembles constructed by replacing ReLU layers with different activation functions, employing various data augmentation techniques, and utilizing several algorithms, including some novel ones, that perturb network weights. Experimental results performed across many data sets representing different tasks demonstrate that our proposed methods for building deep ensembles produces superior results. All the resources required to replicate our experiments are available at https://github.com/LorisNanni.

In this paper we derive a unified perspective for Optimal Transport (OT) theory and Mean Field Control (MFC) theory to analyse the learning process for Neural Networks algorithms in a high-dimensional framework. We consider Mean Field Neural Networks in the context of MFC theory, specifically the mean field formulation of OT theory that allows the development of highly efficient algorithms while providing a powerful tool in the context of explainable Artificial Intelligence.

Precision agriculture is a technology used by farmers to help food sustainability amidst growing population. One of the tools of precision agriculture is yield monitoring, which helps a farmer manage his production. Yield monitoring is usually done during harvest, however it could also be done early in the growing season. Early prediction of yield, specifically for fruit trees, aids the farmer in the marketing of their product and assists in managing production logistics such as labor requirement and storage needs. In this study, a machine vision system is developed to estimate fruit yield early in the season. The machine vision system uses a color camera to capture images of fruit trees during the full bloom period. An image segmentation algorithm based on an artificial neural network was developed to recognize and count the blossoms on the tree. The artificial neural network segmentation algorithm uses color information and position as input. The resulting correlation between the blossom count and the actual number of fruits on the tree shows the potential of this method to be used for early prediction of fruit yield.

With the fast development of the mobile Internet, the online platforms of social networks have rapidly been developing for the purpose of making friends, sharing information, etc. In these online platforms, users being related to each other forms social networks. Literature reviews have shown that social networks have community structure. Through the studies of community structure, the characteristics and functions of networks structure and the dynamical evolution mechanism of networks can be used for predicting user behaviours and controlling information dissemination. Therefore, this study proposes a deep community detection method which includes (1) matrix reconstruction method, (2) spatial feature extraction method and (3) community detection method. The original adjacency matrix in social network is reconstructed based on the opinion leader and nearer neighbors for obtaining spatial proximity matrix. The spatial eigenvector of reconstructed adjacency matrix can be extracted by an auto-encoder based on convolution neural network for the improvement of modularity. In experiments, four open datasets of practical social networks were selected to evaluate the proposed method, and the experimental results show that the proposed deep community detection method obtained higher modularity than other methods. Therefore, the proposed deep community detection method can effectively detect high quality communities in social networks.

This paper presents a study and implementation of a convolutional neural network to identify and recognize humpback whale specimens from the unique patterns of their tails. Starting from a dataset composed of images of whale tails, all the phases of the process of creation and training of a neural network are detailed – from the analysis and pre-processing of images to the elaboration of predictions, using TensorFlow and Keras frameworks. Other possible alternatives are also explained when it comes to tackling this problem and the complications that have arisen during the process of developing this paper.

This report gives a comprehensive summary of two problems about graph convolutional networks (GCNs): over-smoothing and heterophily challenges, and outlines future directions to explore.

Creating intelligent systems capable of recognizing emotions is a difficult task, especially when looking at emotions in animals. This paper describes the process of designing a “proof of concept” system to recognize emotions in horses. This system is formed by two elements, a detector and a model. The detector is a fast region-based convolutional neural network that detects horses in an image. The model is a convolutional neural network that predicts the emotions of those horses. These two elements were trained with multiple images of horses until they achieved high accuracy in their tasks. 400 images of horses were collected and labeled to train both the detector and the model while 80 were used to validate the system. Once the two components were validated, they were combined into a testable system that would detect equine emotions based on established behavioral ethograms indicating emotional affect through head, neck, ear, muzzle and eye position. The system showed an accuracy of between 69% and 74% on the validation set, demonstrating that it is possible to predict emotions in animals using autonomous intelligent systems. Such a system has multiple applications including further studies in the growing field of animal emotions as well as in the veterinary field to determine the physical welfare of horses or other livestock.

: In the last years, materializations of neuromorphic circuits based on nanophotonic arrangements have been proposed, which contain complete optical circuits, laser, photodetectors, photonic crystals, optical fibers, flat waveguides, and other passive optical elements of nanostructured materials, which eliminate the time of simultaneous processing of big groups of data, taking advantage of the quantum perspective and thus highly increasing the potentials of contemporary intelligent computational systems. This article is an effort to record and study the research that has been conducted concerning the methods of development and materialization of neuromorphic circuits of Neural Networks of nanophotonic arrangements. In particular, an investigative study of the methods of developing nanophotonic neuromorphic processors, their originality in neuronic architectural structure, their training methods and their optimization has been realized along with the study of special issues such as optical activation functions and cost functions.

Substances that can modify the androgen receptor pathway in humans and animals are entering the environment and food chain with the proven ability to disrupt hormonal systems and leading to toxicity and adverse effects on reproduction, brain development, and prostate cancer, among others. State-of-the-art databases with experimental data of human, chimp, and rat effects by chemicals have been used to build machine learning classifiers and regressors and evaluate these on independent sets. Different featurizations, algorithms, and protein structures lead to dif- ferent results, with deep neural networks (DNNs) on user-defined physicochemically-relevant features developed for this work outperforming graph convolutional, random forest, and large featurizations. The results show that these user-provided structure-, ligand-, and statistically-based features and specific DNNs provided the best results as determined by AUC (0.87), MCC (0.47), and other metrics and by their interpretability and chemical meaning of the descriptors/features. In addition, the same features in the DNN method performed better than in a multivariate logistic model: validation MCC = 0.468 and training MCC = 0.868 for the present work compared to evalu- ation set MCC = 0.2036 and training set MCC = 0.5364 for the multivariate logistic regression on the full, unbalanced set. Techniques of this type may improve AR and toxicity description and predic- tion, improving assessment and design of compounds. Source code and data are available at https://github.com/AlfonsoTGarcia-Sosa/ML

This work has its origin in intuitive physical and statistical considerations. The problem of optimizing an artificial neural network is treated as a physical system, composed of a conservative vector force field. The derived scalar potential is a measure of the potential energy of the network, a function of the distance between predictions and targets. Starting from some analogies with wave mechanics, the description of the sys-tem is justified with an eigenvalue equation that is a variant of the Schr˜odinger equation, in which the potential is defined by the mutual information between inputs and targets. The weights and parameters of the network, as well as those of the state function, are varied so as to minimize energy, using an equivalent of the variational theorem of wave mechanics. The minimum energy thus obtained implies the principle of minimum mutual information (MinMI). We also propose a definition of the potential work produced by the force field to bring a network from an arbitrary probability distribution to the potential-constrained system, which allows to establish a measure of the complexity of the system. At the end of the discussion we expose a recursive procedure that allows to refine the state function and bypass some initial assumptions, as well as a discussion of some topics in quantum mechanics applied to the formalism, such as the uncertainty principle and the temporal evolution of the system. Results demonstrate how the minimization of energy effectively leads to a decrease in the average error between network predictions and targets.

Speech Emotion Recognition (SER) is an interesting and difficult problem to handle. In this paper, we deal with it through the implementation of deep learning networks. We have designed and implemented six different deep learning networks, a Deep Belief Network (DBN), a simple deep neural network (SDNN), a LSTM network (LSTM), a LSTM network with the addition of an attention mechanism (LSTM-ATN), a Convolutional neural network (CNN), and a Convolutional neural network with the addition of an attention mechanism (CNN-ATN), having in mind, apart from solving the SER problem, to test the impact of attention mechanism to the results. Dropout and Batch Normalization techniques are also used to improve the generalization ability (prevention of overfitting) of the models as well as to speed up the training process. The Surrey Audio-Visual Expressed Emotion database (SAVEE), and the Ryerson Audio-Visual Database (RAVDESS) database were used for training and evaluation of our models. The results showed that networks with the addition of the attention mechanism did better than the others. Furthermore, they showed that CNN-ATN was the best among tested networks, achieving an accuracy of 74% for the SAVEE and 77% for the RAVDESS dataset, and exceeded existing state-of-the-art systems for the same datasets.

This work describes and compares the backpropagation algorithm with the Extended Kalman filter, a second-order training method which can be applied to the problem of learning neural network parameters and is known to converge in only a few iterations. The algorithms are compared with respect to their effectiveness and speed of convergence using simulated data for both, a regression and a classification task.

Graph neural networks (GNNs) build on the success of deep learning models by extending them for use in graph spaces. Transfer learning has proven extremely successful for traditional deep learning problems: resulting in faster training and improved performance. Despite the increasing interest in GNNs and their use cases, there is little research on their transferability. This research demonstrates that transfer learning is effective with GNNs, and describes how source tasks and the choice of GNN impact the ability to learn generalisable knowledge. We perform experiments using real-world and synthetic data within the contexts of node classification and graph classification. To this end, we also provide a general methodology for transfer learning experimentation and present a novel algorithm for generating synthetic graph classification tasks. We compare the performance of GCN, GraphSAGE and GIN across both the synthetic and real-world datasets. Our results demonstrate empirically that GNNs with inductive operations yield statistically significantly improved transfer. Further we show that similarity in community structure between source and target tasks support statistically significant improvements in transfer over and above the use of only the node attributes.

Convolutional Neural Networks (CNNs) are largely hand-crafted, which leads to inefficiency in the constructed network. Various other algorithms have been proposed to address this issue, but the inefficiencies resulting from human intervention have not been addressed. Our proposed EvolveNet algorithm is a task-agnostic evolutionary search algorithm that can find optimal depth and width scales automatically in an efficient way. The optimal configurations are not found using grid search, instead evolved from an existing network. This eliminates inefficiencies that emanate from hand-crafting, thus reducing the drop in accuracy. The proposed algorithm is a framework to search through a large search space of subnetworks until a suitable configuration is found. Extensive experiments on the ImageNet dataset demonstrate the superiority of the proposed method by outperforming the state-of-the-art methods.

Identifying fake news on the media has been an important issue. This is especially true considering the wide spread of rumors on the popular social networks such as Twitter. Various kinds of techniques have been proposed for automatic rumor detection. In this work, we study the application of graph neural networks for rumor classification at a lower level, instead of applying existing neural network architectures to detect rumors. The responses to true rumors and false rumors display distinct characteristics. This suggests that it is essential to capture such interactions in an effective manner for a deep learning network to achieve better rumor detection performance. To this end we present a simplified aggregation graph neural network architecture. Experiments on publicly available Twitter datasets demonstrate that the proposed network has performance on a par with or even better than that of state-of-the-art graph convolutional networks, while significantly reducing the computational complexity.

RBF networks are used in a variety of real-world applications such as medical data or signal processing problems. The success of these parametric models lies in the successful adaptation of their parameters using efficient computational techniques. In the current work, a method of adjusting the parameters of these networks using Grammatical Evolution is presented. Grammatical Evolution will be used to successfully discover the most promising range of parameter values and then the training of the parameter set will be achieved using a Genetic Algorithm. The new method was applied to a wide range of data fitting and classification problems, and the results were more than promising.

The Physics Informed Neural Networks framework is applied to the understanding of the dynamics of Coronavirus of 2019. To provide the governing system of equations used by the framework, the Susceptible-Infected-Recovered-Death mathematical model is used. The study focused on finding the patterns of the dynamics of the disease which involves predicting the infection rate, recovery rate and death rate; thus predicting the active infections, total recovered, susceptible and deceased at any required time. The study used data that was collected on the dynamics of COVID-19 from the Kingdom of Eswatini between March 2020 and September 2021. The obtained results showed less errors thus making highly accurate predictions.

Convolutional neural networks (CNNs), inspired by biological visual cortex systems, are a powerful category of artificial neural networks that can extract the hierarchical features of raw data to greatly reduce the network parametric complexity and enhance the predicting accuracy. They are of significant interest for machine learning tasks such as computer vision, speech recognition, playing board games and medical diagnosis [1-7]. Optical neural networks offer the promise of dramatically accelerating computing speed to overcome the inherent bandwidth bottleneck of electronics. Here, we demonstrate a universal optical vector convolutional accelerator operating beyond 10 Tera-FLOPS (floating point operations per second), generating convolutions of images of 250,000 pixels with 8-bit resolution for 10 kernels simultaneously — enough for facial image recognition. We then use the same hardware to sequentially form a deep optical CNN with ten output neurons, achieving successful recognition of full 10 digits with 900 pixel handwritten digit images with 88% accuracy. Our results are based on simultaneously interleaving temporal, wavelength and spatial dimensions enabled by an integrated microcomb source. This approach is scalable and trainable to much more complex networks for demanding applications such as unmanned vehicle and real-time video recognition.

The accurate prediction of molecular properties such as lipophilicity and aqueous solubility are of great importance and pose challenges in several stages of the drug discovery pipeline. Machine learning methods like graph-based neural networks (GNNs) have shown exceptionally good performance in predicting these properties. In this work, we introduce a novel GNN architecture, called directed edge graph isomorphism network (D-GIN). It is composed of two distinct sub-architectures (D-MPNN, GIN) and achieves an improvement in accuracy over its sub-architectures employing various learning, and featurization strategies. We argue that combining models with different key aspects help make graph neural networks deeper and simultaneously increase their predictive power. Furthermore, we address current limitations in assessment of deep-learning models, namely, comparison of single training run performance metrics, and offer a more robust solution.

Artificial neural networks are a powerful tool for managing data that is difficult to process and interpret. This paper presents the study of artificial neural networks for information processing generated by an optical encoder based on the polarization of light. A machine learning technique is proposed to train the neural networks, such that the system can predict with remarkable accuracy the angular position in which the rotating element of the neuro-encoder is located, based on information provided by light’s phase shifting arrangements. The proposed neural designs show excellent performance in small angular intervals, and a methodology is proposed to avoid losing this remarkable characteristic in measurements from 0 to 180o or even to 360o. The neuro-encoder is implemented in simulation stage to obtain performance results. This study can be useful to improve capabilities of resolvers or other polyphasic sensors.

Speech signals are degraded in real-life environments, product of background noise or other factors. The processing of such signals for voice recognition and voice analysis systems presents important challenges. One of the conditions that make adverse quality difficult to handle in those systems is reverberation, produced by sound wave reflections that travel from the source to the microphone in multiple directions.To enhance signals in such adverse conditions, several deep learning-based methods have been proposed and proven to be effective. Recently, recurrent neural networks, especially those with long and short-term memory (LSTM), have presented surprising results in tasks related to time-dependent processing of signals, such as speech. One of the most challenging aspects of LSTM networks is the high computational cost of the training procedure, which has limited extended experimentation in several cases. In this work, we present a proposal to evaluate the hybrid models of neural networks to learn different reverberation conditions without any previous information. The results show that some combination of LSTM and perceptron layers produce good results in comparison to those from pure LSTM networks, given a fixed number of layers. The evaluation has been made based on quality measurements of the signal's spectrum, training time of the networks and statistical validation of results. Results help to affirm the fact that hybrid networks represent an important solution for speech signal enhancement, with advantages in efficiency, but without a significan drop in quality.

Near Earth Asteroids represent potential threats to human life because their trajectories may bring them in the proximity of the Earth. Monitoring these objects could help predict future impact events, but such efforts are hindered by the large numbers of objects that pass through the Earth’s vicinity. Additionally, there is also the problem of distinguishing asteroids from other objects in the night sky, which implies sifting through large sets of telescope image data. Within this context, we believe that employing machine learning techniques could greatly improve the detection process by sorting out the most likely asteroid candidates to be reviewed by human experts. At the moment, the use of machine learning techniques is still limited in the field of astronomy and the main goal of the present paper is to study the effectiveness of deep CNNs for the classification of astronomical objects, asteroids in this particular case, by comparing some of the well-known deep convolutional neural networks, including InceptionV3, Xception, InceptionResNetV2 and ResNet152V2. We have applied transfer learning and fine-tuning on these pre-existing deep convolutional networks and from the results that we have obtained one can see the potential of using deep convolutional neural networks in the process of asteroid classification. The InceptionV3 model has the best results in the asteroid class, meaning that by using it, we loose the least number of valid asteroids.

A convolutional neural network (CNN) is sometimes understood as a black box in the sense that while it can approximate any function, studying its structure will not give us any insights into the nature of the function being approximated. In other terms, the discriminative ability does not reveal much about the latent representation of a network. This research aims to establish a framework for interpreting the CNNs by profiling them in terms of interpretable visual concepts and verifying them by means of Integrated Gradient. We also ask the question, "Do different input classes have a relationship or are they unrelated?" For instance, could there be an overlapping set of highly active neurons to identify different classes? Could there be a set of neurons that are useful for one input class whereas misleading for a different one? Intuition answers these questions positively, implying the existence of a structured set of neurons inclined to a particular class. Knowing this structure has significant values; it provides a principled way for identifying redundancies across the classes. Here the interpretability profiling has been done by evaluating the correspondence between individual hidden neurons and a set of human-understandable visual semantic concepts. We also propose an integrated gradient-based class-specific relevance mapping approach that takes the spatial position of the region of interest in the input image. Our relevance score verifies the interpretability scores in terms of neurons tuned to a particular concept/class. Further, we perform network ablation and measure the performance of the network based on our approach.

In this paper, the Group Method of Data Handling (GMDH) type of neural networks is used for the inductance calculation of variable inductors. The relation between the inductance of the inductor in the linear and nonlinear regions is investigated, and parameters such as the voltage across the inductor, bias current, and ac current are taken into account. The experimental setup is used for generating the data needed for training the neural network. Over 800 experiments were conducted and were used for training and validation of the neural network results. The results are compared with the reluctance equivalent circuit method, and they show a much better accuracy. The proposed method can be used for the calculation of various magnetic components, and it is not limited to variable inductors.

The principal minimum horizontal stress plays an important role in the study of reservoir characteristics, modeling of oil and gas reservoir, drilling, production and stimulating wells. However, it is currently not possible to measure the minimum horizontal stress along the wellbore as a geophysical parameter logging. Minimum horizontal stress is measured by leak-off test (LOT) at only several points in a well. In order to have the values all along the wellbore, experimental formulas were established to determine the minimum horizontal stresses for different fields. Then these formulas must be calibrated with LOT data whose number is usually limited, even sometimes unavailable. On the other hand, the empirical formulas of one field might not be accurate for another. This study presents a new approach to solve the problem of minimum horizontal stress estimation by a combination of artificial intelligence and geostatistics. The method consists of using artificial neural network (ANN) to build a model of minimum horizontal stress estimation from relevant parameters such as true vertical depth, pore pressure and vertical stress, then combined with Kriging interpolation to obtain the distribution in space of the minimum horizontal stress. Hence, this method can estimate the minimum horizontal stress with a limited amount of available data and therefore we do not need to drill new wells or to find empirical formulas for each survey area. The method was then applied in a case study involved real geomechanical dataset of Hai Thach - Moc Tinh field in Nam Con Son basin, Vietnam.

In the current industrial landscape, increasingly pervaded by technological innovations, the adoption of optimized strategies for asset management is becoming a critical key success factor. Among the various strategies available, the “Prognostics and Health Management” strategy is able to support maintenance management decisions more accurately, through continuous monitoring of equipment health and “Remaining Useful Life” forecasting. In the present study, Convolutional Neural Network-based Deep Neural Network techniques are investigated for the Remaining Useful Life prediction of a punch tool, whose degradation is caused by working surface deformations during the machining process. Surface deformation is determined using a 3D scanning sensor capable of returning point clouds with micrometric accuracy during the operation of the punching machine, avoiding both downtime and human intervention. The 3D point clouds thus obtained are transformed into bidimensional image-type maps, i.e., maps of depths and normal vectors, to fully exploit the potential of convolutional neural networks for extracting features. Such maps are then processed by comparing 15 genetically optimized architectures with the transfer learning of 19 pre-trained models, using a classic machine learning approach, i.e., Support Vector Regression, as a benchmark. The achieved results clearly show that, in this specific case, optimized architectures provide performance far superior (MAPE=0.058) to that of transfer learning which, instead, remains at a lower or slightly higher level (MAPE=0.416) than Support Vector Regression (MAPE=0.857).

All kinds of vessels consist of dozens of complex machineries with rotating parts and electric motors that operate continuously in a harsh environment with excess temperature and humidity, vibration, fatigue and load. A breakdown or malfunction in one of these machineries can significantly impact the vessel’s operation and safety and consequently, the safety of the crew and the environment. To maintain operational efficiency and seaworthiness, the shipping industry invests substantial resources in preventive maintenance and repairs. This research presents the economic and technical benefits of predictive maintenance over to traditional preventive maintenance, and repair by replacement approaches in the maritime domain. By leveraging modern technology and Artificial Intelligence, we can analyze real-time operating conditions of machinery, enabling early detection of potential damages and allowing for effective planning of future maintenance and repair activities. In this paper, we propose and develop a Convolutional Neural Network that is fed with raw vibration measurements acquired in a laboratory environment from the ball bearings of a motor. Then, we investigate whether the proposed network can accurately detect the functional state of ball bearings and categorize any possible failures present, contributing to improved maintenance practices in the shipping industry.

Water resource pollution, particularly in river channels, presents a grave environmental challenge that necessitates a comprehensive and systematic approach encompassing assessment, forecasting, and effective management. This article provides a comprehensive exploration of the methodology and modeling tools employed to scrutinize the process of river channel pollution due to silting, rooted in the fundamental principles of hydrodynamics and pollutant transport dynamics. The study's methodology seamlessly integrates numerical simulations with state-of-the-art neural network techniques, with a specific focus on the Physics-informed neural networks (PINN) method. This innovative approach represents a groundbreaking fusion of artificial neural networks (ANN) and physical equations, offering a more efficient and precise means of modeling a wide array of complex processes and phenomena. The proposed mathematical model, grounded in the Euler equation, has been meticulously implemented using the Ansys Fluent software package, ensuring accuracy and reliability in the computations. In a pivotal phase of the research, a thorough comparative analysis was conducted between the results derived from the PINN method and those obtained through conventional numerical approaches using the Ansys Fluent software package. The outcomes of this analysis revealed the superior performance of the PINN method, characterized by the generation of smoother pressure fluctuation profiles and significantly reduced computation time, underscoring its potential as a transformative modeling tool. The calculated data originating from this study assumes paramount significance in the ongoing battle against river sedimentation. Beyond this immediate application, these findings also serve as a valuable resource for creating predictive materials pertaining to river channel silting, thereby empowering decision-makers and environmental stakeholders with essential information. The utilization of modeling techniques to address pollution concerns in river channels holds the potential to revolutionize risk management and safeguard the integrity of our vital water resources. However, it is imperative to underscore that the effectiveness of such models hinges on ongoing monitoring and frequent data updates, ensuring that they remain aligned with real-world conditions. This research not only contributes to the enhanced understanding and proactive management of river channel pollution due to silting but also underscores the pivotal role of advanced modeling methodologies in the preservation of our invaluable water resources for present and future generations.

In the realm of computer vision, semantic segmentation is the task of recognizing objects in images at the pixel level. This is done by performing a classification of each pixel. The task is complex and requires sophisticated skills and knowledge about the context to identify objects’ boundaries. The importance of semantic segmentation in many domains is undisputed. In medical diagnostics, it simplifies the early detection of pathologies, thus mitigating the possible consequences. In this work, we provide a review of the literature on deep ensemble learning models for polyp segmentation and we develop new ensembles based on convolutional neural networks and transformers. The development of an effective ensemble entails ensuring diversity between its components. To this end, we combine different models (HarDNet-MSEG, Polyp-PVT, and HSNet) trained with different data augmentation techniques, optimization methods, and learning rates, which we experimentally demonstrate to be useful to form a better ensemble. Most importantly, we introduce a new method to obtain the segmentation mask which is more suitable for combining transformers in an ensemble. In our extensive experimental evaluation, the proposed ensembles exhibit state-of-the-art performance.

The Necessity of Human resources beyond perception of human understanding towards the Evaluation of video quality or Data screening methodology is conducted based on human perception level since it is concerned with how visual content is perceived by a observer based on observations with his/ her perception on a particular video sequence. Therefore, we considered that the subject has to grade the encoded video sequences under certain test environment conditions based on ITU-Recommendations. Since Human perception is considered as the true judgment and precise measurement of visual content, data screening has became quite essential and quite comfortable to general public due to introduction of User Experience(UX) concept by User Experience community. The translations of A recurrent neural network is based on certain principles, for instance we considered natural language processing which is certainly adaptable towards understanding sequential data and use patterns to predict the consistency within observers. In our research, we adapted principles based on Recurrent Neural Networks while assuming consistency within observers for predicting video quality within data screening environment towards subjective experiments. Moreover,this research work explores the trade offs between Human perception on visual content and consistency of observations within individual observer.

This article explores the transformative role of neural networks in the realm of biology, particularly within microscopy image processing and illustrations. Neural networks have revolutionized cell segmentation and analysis, enabling precise delineation of cell boundaries and tracking of cellular behaviours. They excel in detecting subcellular structures, unravelling intricate organelle interactions. In 3D image reconstruction, neural networks navigate volumetric datasets, enhancing our understanding of spatial cellular architecture. These networks enhance disease diagnosis by identifying anomalies and irregularities, potentially revolutionizing early detection and classification. Moreover, neural networks restore and enhance microscopy images, unveiling hidden details. Lastly, they bridge art and science, fostering captivating biological art and enriching science communication. As neural networks evolve, they promise a future of limitless possibilities, weaving together technology, science, and art to illuminate the microscopic realm in unprecedented ways.

In the pursuit of energy efficiency and reduced environmental impact, adequate ventilation in enclosed spaces is essential. This study presents a hybrid neural network model designed for real-time monitoring and prediction of environmental variables. The system comprises two phases: An IoT hardware-software platform for data acquisition and decision-making, and a hybrid model combining short-term memory and convolutional recurrent structures. The results are promising and hold potential for integration into parallel processing AI architectures.

This article presents a forecast model that uses a hybrid architecture of recurrent neural networks (RNN) with surface neural networks (ANN), based on historical records of exported active energy (EAE) and weather data. Two types of models were developed: the first type includes six models that use EAE records and weather variables as inputs, while the second type includes eight models that use only weather variables. Different metrics were applied to assess the performance of these models, and the best model of each type was selected. Finally, a comparison of the performance between the selected models of both types is presented, and they are validated with real data provided by a solar plant, achieving acceptable levels of accuracy. The selected model of the first type has an RMSE of 0.19, MSE of 0.03, MAE of 0.09, a correlation coefficient of 0.96, and a determination coefficient of 0.93. The other selected model of the second type showed lower precision in the metrics (RMSE = 0.24, MSE = 0.06, MAE = 0.10, Corr. Coef. = 0.95, and Det. Coef. = 0.90). Both models demonstrated good performance and acceptable accuracy in forecasting the weekly photovoltaic energy production of the solar plant.

Topology optimization is a pioneering design method that can provide various candidates with high mechanical properties. However, the high-resolution for the optimum structures is highly desired, normally in turn leading to computationally intractable puzzle, especially for the famous Solid Isotropic Material with Penalization (SIMP) method. In this paper, an efficient and high-resolution topology optimization method is proposed based on the Super-Resolution Convolutional Neural Network (SRCNN) technique in the framework of SIMP. The SRCNN includes four processes, i.e. refining, path extraction & representation, non-linear mapping, and reconstruction. The high computational efficiency is achieved by a pooling strategy, which can balance the number of finite element analysis (FEA) and the output mesh in optimization process. To further reduce the high computational cost of 3D topology optimization problems, a combined treatment method using 2D SRCNN is built as another speeding-up strategy. A number of typical examples justify that the high-resolution topology optimization method adopting SRCNN has excellent applicability and high efficiency for 2D and 3D problems with arbitrary boundary conditions, any design domain shape, and varied load.

In recent years, the field of deep learning achieved considerable success in pattern recognition, image segmentation and may other classification fields. There are a lot of studies and practical applications of deep learning on images, video or text classification. In this study, we suggest a method for changing the architecture of the most performing CNN models with the aim of designing new models to be used as stand-alone networks or as a component of an ensemble. We propose to replace each activation layer of a CNN (usually a ReLu layer) by a different activation function stochastically drawn from a set of activation functions: in this way the resulting CNN has a different set of activation function layers.

Capability assessment plays a crucial role in the demonstration and construction of equipment. To improve the accuracy and stability of capability assessment, we study the neural network learning algorithms in the field of capability assessment and index sensitivity. Aiming at the problem of over-fitting and parameter optimization in neural network learning, the paper proposes an improved machine learning algorithm—the Ensemble Learning Based on Policy Optimization Neural Networks (ELPONN) algorithm with the policy optimization and ensemble learning. This algorithm presents optimized neural network learning algorithm through different strategies evolution, and builds an ensemble learning model of multi-intelligent algorithms to assessment the capability and analyze the sensitivity of the indexes. Through the assessment of capabilities, the algorithm effectively avoids parameter optimization from entering the minimum point in performance to improve the accuracy of equipment capability assessment, which is significantly better than previous neural network assessment methods. The experimental results show that the mean relative error is 4.10%, which is better than BP, GABP, and early stopping. The ELPONN algorithm has better accuracy and stability performance, and meets the requirements of capability assessment.

In the intricate domain of text analysis, syntactic dependency trees play a crucial role in unraveling the web of relations among entities embedded in the text. These trees provide a structural roadmap, guiding us through the complex syntax to pinpoint the interactions and connections between different entities. However, the challenge lies in sifting through this intricate structure to extract relevant information, a task that requires precision and discernment. Traditional approaches often rely on rule-based pruning methods to simplify these dependency structures, focusing on certain parts while discarding others. Yet, this approach has its pitfalls, as it can overlook critical nuances and connections that are vital for a comprehensive understanding of the text. Addressing this gap, our research introduces the Syntactic Dependency-Aware Neural Networks (SDANNs), a groundbreaking model designed to harness the full power of the entire dependency tree. This approach marks a significant departure from the conventional methods. Instead of the rigid rule-based pruning, SDANNs implement a more flexible and dynamic 'soft-pruning' technique. This method allows the model to adaptively focus on the sub-structures within the dependency tree that are most relevant for understanding the relationships between entities. By doing so, it ensures that no vital information is overlooked, and all potential connections are considered. The efficacy of SDANNs is not just theoretical but has been empirically validated through extensive testing and evaluations across a wide range of tasks. These tasks include the extraction of complex relations spanning multiple sentences, as well as detailed analyses at the sentence level. In each of these scenarios, SDANNs have demonstrated a remarkable ability to leverage the full structural complexity of dependency trees. This capability sets them apart from existing models, enabling a more nuanced and comprehensive analysis of textual relations. The results of these evaluations consistently show that SDANNs not only meet but significantly exceed the performance of prior models. This superiority is evident in the way SDANNs handle the multifaceted and often subtle interactions within the text, offering insights that were previously inaccessible with conventional methods. In summary, the Syntactic Dependency-Aware Neural Networks represent a significant advancement in the field of text analysis. By fully embracing the complexity of syntactic dependency trees and employing a sophisticated 'soft-pruning' approach, SDANNs open new avenues for exploring and understanding the intricate relationships that exist within written language. This model stands as a testament to the potential of combining advanced neural network architectures with a deep understanding of linguistic structures, paving the way for more accurate, nuanced, and comprehensive analyses of text.

Thermal simulations are an important part in the design of electronic systems, especially as systems with high power density become common. In simulation-based design approaches, a considerable amount of time is spent by repeated simulations. In this work, we present a proof-of-concept study of the application of convolutional neural networks to accelerate those thermal simulations. The goal is not to replace standard simulation tools but to provide a method to quickly select promising samples for more detailed investigations. Based on a training set of randomly generated circuits with corresponding Finite Element solutions, the full 3D steady-state temperature field is estimated using a fully convolutional neural network. A custom network architecture is proposed which captures the long-range correlations present in heat conduction problems. We test the network on a separate dataset and find that the mean relative error is around 2 % and the typical evaluation time is 35 ms per sample ( 2 ms for evaluation, 33 ms for data transfer). The benefit of this neural-network-based approach is that, once training is completed, the network can be applied to any system within the design space spanned by the randomised training dataset (which includes different components, material properties, different positioning of components on a PCB, etc.).

Neural networks require a large quantity of training spectra and detector responses in order to learn to solve the inverse problem of neutron spectrum unfolding. In addition, due to the under-determined nature of unfolding, non-physical spectra which would not be encountered in usage should not be included in the training set. While physically realistic training spectra are commonly determined experimentally or generated through Monte Carlo simulation, this can become prohibitively expensive when considering the quantity of spectra needed to effectively train an unfolding network. In this paper, we present three algorithms for the generation of large quantities of realistic and physically motivated neutron energy spectra. Using an IAEA compendium of 251 spectra, we compare the unfolding performance of neural networks trained on spectra from these algorithms, when unfolding real-world spectra, to two baselines. We also investigate general methods for evaluating the performance of and optimizing feature engineering algorithms.

The emergence of the Internet of Things (IoT) has led to the deployment of various types of sensors in many application fields, including environment monitoring, smart cities, health, industries, and others. The increasing number of connected devices has led to the creation of massive quantities of data that need to be analyzed. Typically, this data is ordered by time, as a time series. In this context, this paper presents a time series prediction model based on Recurrent Neural Networks in order to predict one step ahead. Result obtained through five Internet of Things monitoring datasets, showed that the Recurrent Neural Network obtained better performance that the prediction methods, ARIMA and SVM.

Although neural network architectures are critical for their performance, how the structural characteristics of a neural network affect its performance has still not been fully explored. We here map architectures of neural network to directed acyclic graphs, and find that incoherence, a structural characteristic to measure the order of directed acyclic graphs, is a good indicator for the performance of corresponding neural networks. Therefore we propose a deep isotropic neural network architecture by folding a chain of same blocks then connecting the blocks with skip connections at different distances. Our models, named FoldNet, have two distinguishing features compared with traditional residual neural netowrks. First, the distances between block pairs connected by skip connections increase from always equal to one to specially selected different values, which lead to more incoherent graphs and let the neural network explore larger receptive fields and thus enhance its multi-scale representation ability. Second, the number of direct paths increases from one to multiple, which leads to a larger proportion of shorter paths and thus improve the direct propagation of information throughout the entire network. Image classification results on CIFAR-10 and Tiny ImageNet benchmarks suggested that our new network architecture performs better than traditional residual neural networks.

Optical motion capture is a mature contemporary technique for the acquisition of motion data, alas it is non-error-free. Due to technical limitations and occlusions of markers, gaps might occur in such recordings. The article reviews various neural network architectures applied for gap filling problem in motion capture sequences within FBM framework providing the representation for body kinematic structure. The results are compared with interpolation and matrix completion methods. We found out, that for longer sequences simple linear feedforward neural networks can outperform the other, sophisticated architectures. We were also able to identify, that acceleration and monotonicity of input sequence are the parameters that have a notable impact on the obtained results.

We consider the problem of coordinating the charging of an entire fleet of electric vehicles (EV), using a model-free approach, i.e. purely data-driven reinforcement learning (RL). The objective of the RL-based control is to optimize charging actions, while fulfilling all EV charging constraints (e.g. timely completion of the charging). In particular, we focus on batch-mode learning and adopt fitted Q-iteration (FQI). A core component in FQI is approximating the Q-function using a regression technique, from which the policy is derived. Recently, a dueling neural networks architecture was proposed and shown to lead to better policy evaluation in the presence of many similar-valued actions, as applied in a computer game context. The main research contributions of the current paper are that (i)we develop a dueling neural networks approach for the setting of joint coordination of an entire EV fleet, and (ii)we evaluate its performance and compare it to an all-knowing benchmark and an FQI approach using EXTRA trees regression technique, a popular approach currently discussed in EV related works. We present a case study where RL agents are trained with an epsilon-greedy approach for different objectives, (a)cost minimization, and (b)maximization of self-consumption of local renewable energy sources. Our results indicate that RL agents achieve significant cost reductions (70--80%) compared to a business-as-usual scenario without smart charging. Comparing the dueling neural networks regression to EXTRA trees indicates that for our case study's EV fleet parameters and training scenario, the EXTRA trees-based agents achieve higher performance in terms of both lower costs (or higher self-consumption) and stronger robustness, i.e. less variation among trained agents. This suggests that adopting dueling neural networks in this EV setting is not particularly beneficial as opposed to the Atari game context from where this idea originated.

Condition monitoring of rotating shafts is essential for ensuring the reliability and optimal performance of machinery in diverse industries. In this context, as industrial systems become increasingly complex, the need for efficient data processing techniques is paramount. Deep learning has emerged as a dominant approach due to its capacity to capture intricate data patterns and relationships. However, a prevalent challenge lies in the black-box nature of many deep learning algorithms, which often operate without adhering to the underlying physical characteristics intrinsic to the studied phenomena. To address this limitation and enhance the fusion of data-driven methodologies with the fundamental physics of the system under study, this paper leverages physics-informed neural networks (PINNs). Specifically, a simple but realistic numerical case study of an extended Jeffcott rotor model, encompassing damping effects and anisotropic supports for a more comprehensive modelling, is considered. PINNs are used for the estimation of five parameters that characterize the health state of the system. These parameters encompass the radial and angular position of the static unbalance due to the disk installed on the shaft, the stiffness along the principal axes of elasticity, and the non-rotating damping coefficient. The estimation is conducted solely by exploiting the displacement signals from the centre of the disk and, to showcase the efficacy and precision provided by this novel methodology, various scenarios involving different constant rotational speeds are examined. Additionally, the impact of noisy input data is also taken into account within the analysis.

Retinopathy is a human eye disease that causes changes in retinal blood vessels that leads to bleed, leak fluid and vision impairment. Symptoms of retinopathy are blurred vision, changes in color perception, red spots, and eye pain. In this paper, a new methodology based on Convolutional Neural Networks (CNN) is developed and proposed to diagnose and give a decision about the presence of retinopathy. The CNN model is trained by different images of eyes that have retinopathy and those which do not have retinopathy. The performance of the proposed model is compared with the related methods of DREAM, KNN, GD-CNN and SVM. Experimental results show that the proposed CNN performs better.

Abstract Background and Objective: Medical Microwave Radiometry (MWR) is used to capture the thermal properties of internal tissues and has usages in breast cancer detection. Our goal in this paper is to improve classification performance and investigate automated neural architecture search methods. Methods: We investigate optimizing the weights of a weight agnostic neural network using bi-population covariance matrix adaptation evolution strategy (BIPOP-CMA-ES) once the topology is found. We compare it against a weight agnostic and cascade correlation neural network. Results: The experiments are conducted on a breast cancer dataset of 4912 patients. Our proposed weight agnostic BIPOP-CMA-ES model achieved the best performance. It obtained an F1-score of 0.9225, accuracy of 0.9219, precision of 0.9228, recall of 0.9217 and topology of 153 connections. Conclusions: The results are an indication of the potential of MWR utilizing a neural network-based diagnostic tool for cancer detection. By separating the tasks of topology search and weight training, we are able to improve the overall performance.

Over the years, various models have been developed in the stages of the mining process that have allowed predicting and enhancing results, but it is the breakage the variable that connects all the activities of the mining process from the point of view of costs (drilling, blasting, loading, hauling, crushing and grinding). To improve this process, we come up with an idea to develop a breakage prediction model; on the basis of the main variables involved in the drilling and blasting process. For this purpose, we design a computer model based on an Artificial Neuronal Networks (ANN), built by using the most representative variables that come into play with drilling and blasting, such as: the properties of the explosives, the geomechanical parameters of the rock mass, and the design parameters of drilling-blasting. For its experimentation and validation, we have taken the data from a copper mine as reference located in the north of Chile, because of we have the dataset of that ore deposit, which is valid and reliable to evaluate prediction problems based on ANN applied to copper mines. The ANN architecture was of the supervised type, feedforward, with 3 layers and 13 neurons in the only hidden layer, trained with the input data using a dataset with the previously mentioned variables, which then were compared with the breakage results. The model was feed backed in its learning process until it becomes perfected, and is a prediction option that can be used in future blasting of ore deposits with similar characteristics using the same representative variables. Therefore, this is a valid alternative for predicting rock breakage, given that it has been experimentally validated, and has achieved moderately reliable results, providing higher correlation coefficients than traditional models, and with the additional advantage that an ANN model provides, due to its ability to learn and recognize compiled dataset patterns. In this way, using this computer model we can obtain satisfactory results that allow us to predict breakage, providing an alternative for evaluating the costs that this entails.

Optical artificial neural networks (ONNs) have significant potential for ultra-high computing speed and energy efficiency. We report a novel approach to ONNs that uses integrated Kerr optical micro-combs. This approach is programmable and scalable and is capable of reaching ultra-high speeds. We demonstrate the basic building block ONNs — a single neuron perceptron — by mapping synapses onto 49 wavelengths to achieve an operating speed of 11.9 x 10⁹ operations per second, or Giga-OPS, at 8 bits per operation, which equates to 95.2 gigabits/s (Gbps). We test the perceptron on handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. By scaling the perceptron to a deep learning network using off-the-shelf telecom technology we can achieve high throughput operation for matrix multiplication for real-time massive data processing.

This article reviews the introduction of Alzheimer's Disease (AD), neural networks, training and learning of neural networks, applications of neural networks in early diagnosis of AD, applications of neural networks in AD drug discovery, other brain diseases, and challenges faced by AD. First, the paper introduces the background and characteristics of AD. AD is a degenerative neurological disorder characterized by impaired memory, decreased cognitive function, and loss of neurons. These characteristics place a huge burden on the lives and families of patients. Next, the basic principle and structure of neural network are discussed. A neural network is a computational model made up of multiple neurons that can perform tasks by learning and adapting to input data. In particular, the key concepts of neural network hierarchy, activation function and weight adjustment are discussed. Then, the training and learning methods of neural networks are discussed. Common techniques such as backpropagation algorithm and gradient descent optimizer are introduced in detail, as well as the importance of data preprocessing and model evaluation. Next, the paper focuses on the application of neural network in early diagnosis of AD. By extracting features from brain image data, neural networks can automatically identify differences between AD patients and healthy subjects, enabling early diagnosis and intervention. In addition, the application of neural networks in AD drug discovery is also discussed. By analyzing and predicting a database of known drugs, neural networks can help discover potential treatments for AD and speed up the drug discovery process. The paper further explores the application of neural networks in other brain diseases and highlights the challenges faced by AD, such as the lack of reliable biomarkers, complex pathological mechanisms, etc. In summary, this paper presents a systematic overview of AD, neural networks, training and learning of neural networks, applications of neural networks in early diagnosis of AD and drug discovery, and other brain diseases and challenges associated with AD.

State-of-the-art remote sensing scene classification methods employ different Convolutional Neural Network architectures for achieving very high classification performance. A trait shared by the majority of these methods is that the class associated with each example is ascertained by examining the activations of the last fully connected layer, and the networks are trained to minimize the cross-entropy between predictions extracted from this layer and ground-truth annotations. In this work, we extend this paradigm by introducing an additional output branch which maps the inputs to low dimensional representations, effectively extracting additional feature representations of the inputs. The proposed model imposes additional distance constrains on these representations with respect to identified class representatives, in addition to the traditional categorical cross-entropy between predictions and ground-truth. By extending the typical cross-entropy loss function with a distance learning function, our proposed approach achieves significant gains across a wide set of benchmark datasets in terms of classification, while providing additional evidence related to class membership and classification confidence.

This paper presents a study of an automated system for identifying planktic foraminifera at the species level. The system uses a combination of deep learning methods, specifically Convolutional Neural Networks (CNNs), to analyze digital images of foraminifera taken at different illumination angles. The dataset is composed of 1437 groups of sixteen grayscale images, one group for each foraminifer, that are then converted to RGB images with various processing methods. These RGB images are fed into a set of CNNs, organized in an Ensemble Learning (EL) environment. The ensemble is built by training different networks using different approaches for creating the RGB images. The study finds that an ensemble of CNN models trained on different RGB images improves the system's performance compared to other state-of-the-art approaches. The proposed system was also found to outperform human experts in classification accuracy.

Formaldehyde (HCHO) is one of the most important carcinogenic air contaminants. However, the lack of global surface concentration of HCHO monitoring is currently hindering research on outdoor HCHO pollution. Traditional methods are either restricted to small areas or data- demanding for a global scale of research. To alleviate this issue, we adopted neural networks to estimate surface HCHO concentration with confidence intervals in 2019, where HCHO vertical column density data from TROPOMI, in-situ data from HAPs (harmful air pollutants) monitoring network and ATom mission are utilized. Our result shows that the global surface HCHO average concentration is 2.30 μg/m3. Furthermore, in terms of regions, the concentration in Amazon Basin, Northern China, South-east Asia, Bay of Bengal, Central and Western Africa are among the highest. The results from our study provides a first dataset of the global surface HCHO concentration. In addition, the derived confidence interval of surface HCHO concentration adds an extra layer for the confidence to our results. As a pioneer work in adopting confidence interval estimation into AI-driven atmospheric pollutant research and the first global HCHO surface distribution dataset, our paper will pave the way for the rigorous study on global ambient HCHO health risk and economic loss, thus providing a basis for pollutant controlling policies worldwide.

The Physics-Informed Neural Networks (PINNs) improve the efficiency of data utilization by combining physical principles with neural network algorithms and ensure that the predictions are consistent and stable with the physical laws. PINNs opens up a new approach to address inverse problems in fluid mechanics. Based on the single-relaxation-time lattice Boltzmann method (SRT-LBM) with the Bhatnagar-Gross-Krook (BGK) collision operator, the PINN-SRT-LBM model is proposed in this paper for solving the inverse problem in fluid mechanics. The PINN-SRT-LBM model consists of three components. The first component involves a deep neural network that predicts the equilibrium control equations in different discrete velocity directions within SRT-LBM. The second component employs another deep neural network to predict non-equilibrium control equations, enabling inference of the fluid's non-equilibrium characteristics. The third component, a physics informed function translates the outputs of the first two networks into physical infor-mation. By minimizing the residuals of the physical partial differential equations (PDEs), the physics informed function infers relevant macroscopic quantities of the flow. The model evolves two sub-models applicable to different dimensions, named PINN-SRT-LBM-I and PINN-SRT-LBM-II models according to the construction of the physical informed function. The innovation of this work is the introduction of SRT-LBM and discrete velocity models as physical drivers into the neural network through the interpretation function. Therefore, PINN-SRT-LBM allows the neural network to handle inverse problems of various dimensions and focus on problem-specific solving. Experimental results confirm the accurate prediction of flow infor-mation at different Reynolds numbers within the computational domain. Relying on the PINN-SRT-LBM models, inverse problems in fluid mechanics can be solved efficiently.

To drive safely, the driver must be aware of the surroundings, pay attention to the road traffic, and be ready to adapt to new circumstances. Most studies on driving safety focus on detecting anomalies in driver behavior and monitoring the cognitive capabilities of drivers. In our study, we proposed a classifier for basic activities in driving a car, based on a similar approach that could be applied to the recognition of basic activities in daily life, that is, using electrooculographic (EOG) signals and a one-dimensional convolutional neural network (1D CNN). Our classifier achieved an accuracy of 80% for the 16 primary and secondary activities. The accuracy related to primary activities in driving, including crossroad, parking, roundabout was 97.9%, 96.8%, 97.4%, and 99.5%, respectively. The F1 score for secondary driving actions (0.99) was higher than for primary driving activities (0.93–0.94). Furthermore, using the same algorithm, it was possible to distinguish four secondary activities related to activities of daily life and secondary when driving a car.

Singing melody extraction is the task that identifies the melody pitch contour of singing voice from polyphonic music. Most of the traditional melody extraction algorithms are based on calculating salient pitch candidates or separating the melody source from the mixture. Recently, classification-based approach based on deep learning has drawn much attentions. In this paper, we present a classification-based singing melody extraction model using deep convolutional neural networks. The proposed model consists of a singing pitch extractor (SPE) and a singing voice activity detector (SVAD). The SPE is trained to predict a high-resolution pitch label of singing voice from a short segment of spectrogram. This allows the model to predict highly continuous curves. The melody contour is smoothed further by post-processing the output of the melody extractor. The SVAD is trained to determine if a long segment of mel-spectrogram contains a singing voice. This often produces voice false alarm errors around the boundary of singing segments. We reduced them by exploiting the output of the SPE. Finally, we evaluate the proposed melody extraction model on several public datasets. The results show that the proposed model is comparable to state-of-the-art algorithms.

In Holter monitoring, the precise detection of standard heartbeats and Ventricular Premature Contractions (PVCs) is paramount for accurate cardiac rhythm assessment. This study introduces a novel application of the 1D U-Net neural network architecture aimed at enhancing PVC detection in Holter recordings. Training data comprised the Icentia11k, INCART DB, and our custom dataset. The model’s efficacy was subsequently validated against traditional Holter analysis methodologies across multiple databases, including AHA DB, MIT 11 DB, NST, and another custom dataset encompassing challenging real-world examples. The results underscored the 1D U-Net model’s prowess in QRS complex detection, achieving near-perfect balanced accuracy scores across all databases. PVC detection exhibited variability, with balanced accuracy scores ranging from 0.909 to 0.986. Despite some databases, like the AHA DB, showcasing lower sensitivity metrics, their robust balanced accuracy accentuates the model’s equitable performance in discerning both false positives and negatives. In conclusion, while the 1D U-Net architecture is a formidable tool for QRS detection, there’s a clear avenue for further refinement in its PVC detection capability, given the inherent complexities and noise challenges in real-world PVC occurrences.

This paper presents the use of Graph Neural Networks (GNNs) to predict the tensile strength of Fused Deposition Modeling (FDM) specimens. In the present work, there are four main input parameters i.e. Infill percentage, Layer height, Print speed and Extrusion temperature while the Tensile Strength is an output parameter were considered. This study includes use of central composite design based response surface methodology to finalize the experimental layout. 3D printed specimen were manufactured as per the ASTM E8 standard on FDM printer using Polylactic Acid (PLA) as filament. Micro-tensile test were performed on the printed specimen as per ASTM E8 standard. The GNN algorithm was trained on a dataset of FDM specimens, achieving a mean squared error (MSE) of 2.47, mean absolute error (MAE) of 1.14, and R-squared value of 0.78. An adjacency matrix, which shows the connections between nodes in a graph. The obtained plot for nodes and weights in a GNN provide valuable information about the model and its performance. The results show the potential of using GNNs in predicting the mechanical properties of additively manufactured specimens and provide a promising direction for further research in this field.

In urban Vehicular Ad Hoc Network (VANET) environments, buildings play a crucial role as they can act as obstacles that attenuate the transmission signal between vehicles. Quantifying the impact of buildings on the transmission quality is essential, especially in critical scenarios involving emergency vehicles, where reliable communication is of utmost importance. In this research, we propose a supervised learning approach based on artificial neural networks (ANNs) to develop a predictive model capable of estimating the level of signal degradation, represented by the bit error rate (BER), based on the obstacles perceived by moving emergency vehicles. By establishing a relationship between the level of signal degradation and the encountered obstacles, our proposed mechanism enables efficient routing decisions to be made prior to the transmission process. Consequently, data packets are routed through paths that exhibit the lowest BER. To gather the necessary training data, we employed SUMO and NS-3 simulations. The simulation results demonstrate that our developed model successfully learns and accurately estimates the BER for new data instances. Overall, our research contributes to enhancing the performance and reliability of communication in urban VANET environments, especially in critical scenarios involving emergency vehicles, by leveraging supervised learning and artificial neural networks to predict signal degradation levels and optimize routing decisions accordingly.

Over the last several decades, large wildfires are increasingly common across the United States causing disproportionate impact on forest health and function, human well-being, and economy. Here, we examine the severity of large wildfires across the Contiguous United States over the past decade (2011-2020) using a wide array of meteorological, vegetational, and topographical features in the Deep Neural Network model. A total of 4,538 wildfire incidents were used in the analysis covering 87,305 square miles of burned area. We observed the highest number of large wildfires in California, Texas, and Idaho, with lightning causing 43 % of these incidents. Importantly, results indicate that the severity of wildfire occurrences is highly correlated with the climatological forcings, land cover, location, and elevation of the ecosystem. Overall, results may serve useful guide in managing landscapes under changing climate and disturbance regimes.

Recently, various relations and criteria have been presented to establish a proper relationship between control systems and control Global Positioning System (GPS)-intelligent buoy system. Given the importance of controlling the position of buoys and the construction of intelligent systems, in this paper, dynamic system modeling is applied to position marine buoys through the improved neural network with a backstepping technique. This study aims at developing a novel controller based on adaptive fuzzy neural network to optimally track the dynamically positioned vehicle on water with unavailable velocities and unidentified control parameters. In order to model the network with the proposed technique, uncertainties and the unwanted disturbances are studied in the neural network. The presented study aims at developing a neural controlling which applies the vectorial back-stepping technique to the surface ships, which have been dynamically positioned with undetermined disturbances and ambivalences. Moreover, the objective function is to minimize the output error for the neural network (NN) based on closed-loop system. The most important feature of the proposed model for the positioning buoys is its independence from comparative knowledge or information on the dynamics and the unwanted disturbances of ships. The numerical and obtained consequences demonstrate that the controller system can adjust the routes and the position of the buoys to the desired objective with relatively few position errors.

Boiler efficiency is called to some extent of total thermal energy which can be recovered from the fuel. Boiler efficiency losses are due to four major factors: the dry gas flux, the latent heat of steam in the flue gas, the combustion loss or the loss of unburned fuel, radiation and convection losses. In this research, the thermal behavior of boilers in gas refinery facilities is studied and their efficiency and their losses are calculated. The main part of this research is comprised of analyzing the effect of various parameters on efficiency such as excess air, fuel moisture, air humidity, fuel and air temperature, the temperature of combustion gases, and thermal value of the fuel. Based on the obtained results, it is possible to analyze and make recommendations for optimizing boilers in the gas refinery complex using response-surface method (RSM).

This paper inquires on the options pricing modeling using Artificial Neural Networks to price Apple(AAPL) European Call Options. Our model is based on the premise that Artificial Neural Networks can be used as functional approximators and can be used as an alternative to the numerical methods to some extent, for a faster and an efficient solution. This paper provides a neural network solution for two financial models, the BlackScholes-Merton model, and the calibrated-Heston Stochastic Volatility Model, we evaluate our predictions using the existing numerical solutions for the same, the analytic solution for the Black-Scholes equation, COS-Model for Heston’s Stochastic Volatility Model and Standard Heston-Quasi analytic formula. The aim of this study is to find a viable time-efficient alternative to existing quantitative models for option pricing.

This paper analyzes the robust dissipativity of uncertain stochastic generalized neural networks (USGNNs) with Markovian jumping parameters and time-varying delays. In practical applications most of the systems refer to uncertainties, hence, the norm-bounded parameter uncertainties and stochastic disturbance are considered. Then, by constructing an appropriate Lyapunov-Krasovskii functional (LKF) and by employing integral inequalities LMI-based sufficient conditions of the considered systems are established. Numerical simulations are given to show the merit of the presented results.

The paper illustrates a general framework in which a neural network application can be easily integrated and proposes a traffic forecasting approach that uses neural networks based on graphs. The method minimizes the communication network (between vehicles and the database servers) load and represents a reasonable trade-off between communication network load and forecasting accuracy. Traffic prediction leads to the choice of less congested routes and therefore to the reduction of energy consumption. The traffic is forecasting using a LTSM neural network with a regression layer. The inputs of the neural network are sequences - obtained from graph that represent the road network - at specific moments of time that are read from traffic sensors or the outputs of neural network (forecasting sequences). The input sequences can be filtered to improve the forecasting accuracy. This general framework is based on Contiki IoT operating system that ensure support for wireless communication and efficient implementation of processes in a resource constrained system and it is particularized to implement a graph neural network. Two cases are studied: one case in which the traffic sensors are periodically read and the other case in which the traffic sensors are read when their values changes are detected. A comparison between the cases is made and the influence of filtering is evaluated. The obtained accuracy is very good, very close to the accuracy obtained in infinite precision simulation, and the computation time is low enough and the system can work in real time.

Next-generation cancer and oncology research needs to take full advantage of the multi-modal structured, or graph, information, with the graph datatypes ranging from molecular structures to spatially resolved imaging and digital pathology to biological networks to knowledge graphs. Graph Neural Networks (GNNs) efficiently combine the graph structure representations with the high predictive performance of deep learning, especially on the large multi-modal datasets. In this review article, we survey the landscape of recent (2020-present) GNN applications in the context of cancer and oncology research, and delineate six currently predominant research areas. Subsequently, we identify the most promising directions for future research. We compare GNNs with graphical models and "non-structured" deep learning, and devise the guidelines for cancer and oncology researchers or physician-scientists asking the question of whether they should adopt the GNN methodology in their research pipelines.

In South Korea, demolition waste (DW) management has become increasingly significant owing to the rising number of old buildings. Effective DW management requires an efficient approach that accurately quantifies and predicts the generation of DW (DWG) of various types, which necessitates access to the required information or technology capable of achieving this. Hence, we developed an artificial intelligence-based model that predicts the generation of ten DW types, specifically from buildings in redevelopment areas. We used an artificial neural network algorithm with < 10 neurons in the hidden layer to derive individual input variables and optimal hyperparameters for each DW type. All DWG prediction models achieved an average validation and test prediction performance (R²) of 0.970 and 0.952, respectively, with their ratios of percent deviation ≥ 2.5, ver-ifying them as excellent models. Moreover, a Shapley additive explanations analysis revealed that DWG was most impacted by the floor area for all the DW types, with a positive correlation with DWG. Conversely, other factors showed either a positive or negative correlation with DWG de-pending on the DW type. The study findings will enable demolition companies and local gov-ernments in making informed decisions for efficient DW management and resource allocation by accurately predicting the generation of various types of DW.

The continuous and precise mapping of glacier calving fronts is essential for monitoring and understanding rapid glacier changes in Antarctica and Greenland, which have the potential for significant sea level rise within the current century. This effort has been mostly restricted to the slow and painstaking manual digitalization of the calving front positions in thousands of satellite imagery products. Here, we have developed a machine learning toolkit to robustly and automatically detect glacier calving front margins in satellite imagery. The toolkit is based on semantic image segmentation using Convolutional Neural Networks (CNN) with a modified U-Net architecture to isolate the calving fronts from satellite images after having been trained with a dataset of images and their corresponding manually-determined calving fronts. As a case study we train our neural network on a varied set Landsat images with lowered resolutions from Jakobshavn, Sverdrup, and Kangerlussuaq glaciers, Greenland and test the results on novel images from Helheim glacier, Greenland to evaluate the performance of the approach. The neural network is able to identify the calving front in new images with a mean deviation of 96.3 m from the true fronts, equivalent to 1.97 pixels on average, while the corresponding error for manually-determined fronts on the same resolution images is 92.5 m. We find that the trained neural network significantly outperforms common edge detection techniques, and can be used to continuously map out calving-ice fronts with a variety of data products.

This paper reviews non-intrusive load monitoring (NILM) approaches that employ deep neural networks to disaggregate appliances from low frequency data, i.e. data with sampling rates lower than the AC base frequency. We first review the many degrees of freedom of these approaches, what has already been done in literature, and compile the main characteristics of the reviewed publications in an extensive overview table. The second part of the paper discusses selected aspects of the literature and corresponding research gaps. In particular, we do a performance comparison with respect to reported MAE and F$_1$-scores and observe different recurring elements in the best performing approaches, namely data sampling intervals below 10\,s, a large field of view, the usage of GAN losses, multi-task learning, and post-processing. Subsequently, multiple input features, multi-task learning and related research gaps are discussed, the need for comparative studies is highlighted, and finally, missing elements for a successful deployment of NILM approaches based on deep neural networks are pointed out. We conclude the review with an outlook on possible future scenarios.

In the last decade, unprecedented progress in the development of neural networks influenced dozens of different industries, among which are signal processing for the electroencephalography process (EEG). Electroencephalography, even though it appeared in the first half of the 20th century, to this day didn’t change the physical principles of operation. But the signal processing technique due to the use of neural networks progressed significantly in this area. Evidence for this can serve that for the past 5 years more than 1000 publications on the topic of using machine learning have been published in popular libraries. Many different models of neural networks complicate the process of understanding the real situation in this area. In this manuscript, we provided the most comprehensive overview of research where were used neural networks for EEG signal processing.

With the exponential growth of machine learning and development of Artificial Neural Network (ANNs) in recent years, there is great opportunity to leverage this approach and accelarate bio-logical discoveries through applications on the analysis of bioinformatics data. Various types of datasets including for example protein or gene interaction networks, molecular structures and cellular signalling pathways, have already been used for machine learning by training ANNs for inference and pattern classification. However, unlike regular data structures that are commonly used in the computer science and engineering fields, bioinformatics datasets present challenges that require unique algorithmic approaches. The recent development of the geometric and deep learning approach within the machine learning field, is very promising towards accelerating analysis complex bioinformatics datasets. The principles of ANNs and their importance for bio-informatics machine learning is demonstrated herein, through presentation of the undelying mathematical and statistical foundations from group theory, symmetry, linear algebra. Further-more, the structure and functions of ANN algorithms that form the core principles of artificial intelligence are explained, in relation to the bioinformatics data domain. Overall, the manuscript provides guidance for researchers to understand the principles required for practicing machine learning and artificial intelligence, with the special considerations towards bioinformatics applications.

With the widespread of embedded sensing capabilities of mobile devices, there has been unprecedented development of context-aware solutions. This allows the proliferation of various intelligent applications such as those for remote health and lifestyle monitoring, intelligent personalized services, etc. However, activity context recognition based on multivariate time series signals obtained from mobile devices in unconstrained conditions is naturally prone to imbalance class problems. This means that recognition models tend to predict classes with the majority number of samples whilst ignoring classes with the least number of samples, resulting in poor generalization. To address this problem, we propose to augment the time series signals from inertia sensors with signals from ambient sensing to train deep convolutional neural networks (DCNN) models. DCNN provides the characteristics that capture local dependency and scale invariance of these combined sensor signals. Consequently, we developed a DCNN model using only inertial sensor signals and then developed another model that combined signals from both inertia and ambient sensors aiming to investigate the class imbalance problem by improving the performance of the recognition model. Evaluation and analysis of the proposed system using data with imbalanced classes show that the system achieved better recognition accuracy when data from inertial sensors are combined with those from ambient sensors such as environment noise level and illumination, with an overall improvement of 5.3% accuracy.

When extract building from high resolution remote sensing image with meter/sub-meter accuracy, the shade of trees and interference of roads are the main factors of reducing the extraction accuracy. Proposed a Bayesian Convolutional Neural Networks(BCNET) model base on standard fully convolutional networks(FCN) to solve these problems. First take building with no shade or artificial removal of shade as Sample-A, woodland as Sample-B, road as Sample-C. Set up 3 sample libraries. Learn these sample libraries respectively, get their own set of feature vector; Mixture Gauss model these feature vector set, evaluate the conditional probability density function of mixture of noise object and roofs; Improve the standard FCN from the 2 aspect:(1) Introduce atrous convolution. (2) Take conditional probability density function as the activation function of the last convolution. Carry out experiment using unmanned aerial vehicle(UVA) image, the results show that BCNET model can effectively eliminate the influence of trees and roads, the building extraction accuracy can reach 97%.

Direct measurement of strain within muscle is important for understanding muscle function in health and disease. Current technology (kinematics, dynamometry, electromyography) provides limited ability to measure strain within muscle. Regional fiber orientation and length are related with active/passive strain within muscle. Currently, ultrasound imaging provides the only non-invasive means of observing regional fiber orientation within muscle during dynamic tasks. Previous attempts to automatically estimate fiber orientation from ultrasound are not adequate, often requiring manual region selection, feature engineering, providing low-resolution estimations (one angle per muscle), and deep muscles are often not attempted. Here, we propose deconvolutional neural networks (DCNN) for estimating fiber orientation at the pixel-level. Dynamic ultrasound images sequences of the calf muscles were acquired (25 Hz) from 8 healthy volunteers (4 male, ages: 25–36, median 30). A combination of expert annotation and interpolation/extrapolation provided labels of regional fiber orientation for each image. We then trained DCNNs both with and without dropout using leave one out cross-validation. Our results demonstrated robust estimation of regional fiber orientation with approximately 3° error, which was an improvement on previous methods. The methods presented here provide new potential to study muscle in disease and health.

Autonomous target recognition (ATR) plays a crucial role in maintaining situational awareness during environmental monitoring. Unmanned aerial vehicles (UAVs) equipped with autonomous target recognition technology can gather and analyze real-time information about targets, including their locations, sizes, and types. However, UAV-captured images in complex real-world environments often display significant variations in perspective and scale due to changes in UAV altitude and distance. Existing methods for autonomous target recognition on UAVs struggle to capture targets from large field-of-view and multi-scale images, resulting in low recognition accuracy and high false-positive rates. This paper introduces two novel Slimmable neural network models, namely SE-YOLOv5s and ST-YOLOv5s, which are based on the YOLOv5s architecture. These models incorporate the Squeeze and Excitation and Swin-Transformer mechanisms to enhance the ability to extract features from large field-of-view and multi-scale images. To evaluate their performance, experiments were conducted on the Visdrone19 aerial dataset. Compared to the state-of-the-art YOLOv5s algorithm, the utilization of SE-YOLOv5s and ST-YOLOv5s for autonomous target recognition on low-altitude drones resulted in improvements in both accuracy and false-positive rates. These proposed methods combine Slimmable neural network design with feature enhancement mechanisms, addressing the challenges posed by complex real-world environments in UAV missions. The advancements in autonomous target recognition on low-altitude drones will significantly contribute to enhancing situational awareness in future environmental monitoring.

In this research, we present a comprehensive methodology to categorize perfumes based on their fragrance profiles and subsequently aid in creating innovative odoriferous molecules using advanced neural networks. Drawing from data on Parfumo and the Good Scents Company webpage (Parfumo, 2008; The Good Scents Company, 2021), the study employs sophisticated web scraping techniques to gather diverse perfume attributes. Following this, a k-means algorithm is applied for perfume clustering, paving the way for recommending similar scents to consumers. The process then bridges customer preferences to molecular design by incorporating their feedback into generating new molecules via graph neural networks (GNNs). Through converting the Simple Molecular Input Line Entry System (SMILES) representation into graph structures, the GNN facilitates the creation of new molecular designs attuned to consumer desires. The proposed approach offers promising avenues for consumers to pinpoint similar perfume choices, incorporating feedback, and for manufacturers to conceptualize new fragrant molecules with a high likelihood of market resonance.

In this paper, we present a groundbreaking methodology in the realm of aspect-level sentiment analysis, which capitalizes on the advanced capabilities of graph-based neural networks. Our approach, distinguished as the Aspect Correlation Graph Network (ACGN), represents a significant departure from conventional models. These traditional models often analyze aspects in isolation, failing to capture the intricate web of sentiment relationships that may exist within a single sentence. ACGN, however, is designed to address this gap by employing a sophisticated bidirectional attention mechanism, integrated with positional encoding. This unique combination not only enhances the model's ability to focus on relevant parts of the sentence but also aids in constructing detailed, aspect-focused representations. These representations are particularly crucial for understanding the nuanced interplay of sentiments associated with different aspects. Central to our model's architecture is the incorporation of a graph convolutional network. This network serves as a pivotal component in mapping and analyzing the complex network of sentiment correlations that can exist among various aspects within sentences. Through this integration, ACGN is able to unearth and interpret the subtle and often overlooked sentiment dynamics that traditional models might miss. Our comprehensive evaluations of the Aspect Correlation Graph Network, conducted using the SemEval 2014 datasets, have yielded promising results. These findings demonstrate a clear and significant advancement over the capabilities of existing models. Particularly, the results underscore the critical importance and utility of recognizing and understanding the connections between sentiments of different aspects in text analysis. This insight opens new avenues in the field of sentiment analysis, suggesting a broader application potential of ACGN in various contexts where understanding nuanced sentiment relationships is key. Overall, our study not only introduces a novel approach in aspect-level sentiment analysis but also sets a new standard for future research in this area. By highlighting the integral role of inter-aspect sentiment connections, ACGN paves the way for more sophisticated and accurate sentiment analysis tools, capable of handling the complexities of natural language with greater finesse and precision.

Alzheimer's disease (AD) is a progressive and evolving neurodegenerative disease with an insidious onset that can lead to memory loss and cognitive impairment. There is no effective treatment for this disease. However, early diagnosis plays an important role in treatment planning to slow down its progression, as treatment has the greatest impact in the early stages of the disease. Neurological images obtained through different imaging techniques provide powerful information and help diagnose the disease. With the wide application of deep learning techniques in disease diagnosis, especially the prominence of Convolutional Neural Networks (CNNs) in computer vision and image processing, more and more studies are proposing the use of this algorithm for the diagnosis of AD. In this paper, we first systematically introduce the impact of AD on people, detailing the biomarkers, early clinical symptoms, and risk factors of this disease. Secondly, it goes on to detail the development of CNNs, their form, and methods to help diagnose AD. It is proposed that CNNs can help diagnose AD by analyzing medical imaging data, particularly structural brain scans such as magnetic resonance imaging (MRI) and functional scans such as positron emission tomography (PET). Finally, it is concluded that CNNs are of great importance for the diagnosis of AD and that they are likely to play an increasingly important role in the early detection of the disease, the understanding of disease mechanisms, and ultimately, in the development of effective AD therapies and interventions. CNNs are playing an increasingly important role in the Their potential impact on healthcare emphasizes the importance of continued research and innovation in neural networks and medical imaging.

Hopfield Neural Networks (HNNs) are recurrent neural networks used to implement associative memory. Their main feature is their ability to pattern recognition, optimization, or image segmentation. However, sometimes it is not easy to provide the users with good explanations about the results obtained with them due to mainly the large number of changes in the state of neurons (and their weights) produced during a problem of machine learning. There are currently limited techniques to visualize, verbalize, or abstract HNNs. This paper outlines how we can construct automatic video generation systems to explain their execution. This work constitutes a novel approach to get explainable artificial intelligence systems in general and HNNs in particular building on the theory of data-to-text systems and software visualization approaches. We present a complete methodology to build these kinds of systems. Software architecture is also designed, implemented, and tested. Technical details about the implementation are also detailed and explained. Finally, we apply our approach for creating a complete explainer video about the execution of HNNs on a small recognition problem.

The small satellite market continues to grow year after year. A compound annual growth rate of 17% is estimated during the period between 2020 and 2025. Low-cost satellites can send a vast amount of images to be post-processed at the ground to improve the quality and extract detailed information. In this domain lies the resolution enhancement task, where a low-resolution image is converted to a higher resolution automatically. Deep learning approaches to Super-Resolution (SR) reached the state-of-the-art in multiple benchmarks; however, most of them were studied in a single-frame fashion. With satellite imagery, multi-frame images can be obtained at different conditions giving the possibility to add more information per image and improve the final analysis. In this context, we developed and applied to the PROBA-V dataset of multi-frame satellite images a model that recently topped the European Space Agency’s Multi-frame Super Resolution (MFSR) competition. The model is based on proven methods that worked on 2D images tweaked to work on 3D: the Wide Activation Super Resolution (WDSR) family. We show that with a simple 3D CNN residual architecture with WDSR blocks and a frame permutation technique as data augmentation better scores can be achieved than with more complex models. Moreover, the model requires few hardware resources, both for training and evaluation, so it can be applied directly from a personal laptop.

The COVID-19 pandemic situation has created even more difficulties in the quick identification and screening of the COVID-19 patients for the medical specialists. Therefore, a significant study is necessary for detecting COVID-19 cases using an automated diagnosis method, which can aid in controlling the spreading of the virus. In this paper, the study suggests a Deep Convolutional Neural Network-based multi-classification approach (COV-MCNet) using eight different pre-trained architectures such as VGG16, VGG19, ResNet50V2, DenseNet201, InceptionV3, MobileNet, InceptionResNetV2, Xception which are trained and tested on the X-ray images of COVID-19, Normal, Viral Pneumonia, and Bacterial Pneumonia. The results from 3-class (Normal vs. COVID-19 vs. Viral Pneumonia) showed that only the ResNet50V2 model provides the highest classification performance (accuracy: 95.83%, precision: 96.12%, recall: 96.11%, F1-score: 96.11%, specificity: 97.84%) compared to rest of the models. The results from 4-class (Normal vs. COVID-19 vs. Viral Pneumonia vs. Bacterial Pneumonia) demonstrated that the pre-trained model DenseNet201 provides the highest classification performance (accuracy: 92.54%, precision: 93.05%, recall: 92.81%, F1-score: 92.83%, specificity: 97.47%). Notably, the ResNet50V2 (3-class) and DenseNet201 (4-class) models in the proposed COV-MCNet framework showed higher accuracy compared to the rest six models. This indicates that the designed system can produce promising results to detect the COVID-19 cases on the availability of more data. The proposed multi-classification network (COV-MCNet) significantly speeds up the existing radiology-based method, which will be helpful to the medical community and clinical specialists for early diagnosis of the COVID-19 cases during this pandemic.

Clients are increasingly looking for fast and effective means to quickly and frequently survey and communicate the condition of their buildings so that essential repairs and maintenance work can be done in a proactive and timely manner before it becomes too dangerous and expensive. Traditional methods for this type of work commonly comprise of engaging building surveyors to undertake a condition assessment which involves a lengthy site inspection to produce a systematic recording of the physical condition of the building elements, including cost estimates of immediate and projected long-term costs of renewal, repair and maintenance of the building. Current asset condition assessment procedures are extensively time consuming, laborious, and expensive and pose health and safety threats to surveyors, particularly at height and roof levels which are difficult to access. We propose a method for automated detection and localisation of key building defects from images using deep learning and convolution neural networks. The proposed model is based on a pre-trained VGG-16 classifier with Class Activation Mapping (CAM) for object localisation. The model has proven to be robust and able to accurately detect and localise mould growth, stains, and paint deterioration defects arising from dampness in buildings. The approach is being developed with potentials to scale-up to support automated detection of defects and deterioration of buildings in real-time using mobile devices and drones.

Polar mesocyclones (MCs) are small marine atmospheric vortices. The class of intense MCs, called polar lows, are accompanied by extremely strong surface winds and heat fluxes and thus largely influencing deep ocean water formation in the polar regions. Accurate detection of polar mesocyclones in high-resolution satellite data, while challenging, is a time-consuming task, when performed manually. Existing algorithms for the automatic detection of polar mesocyclones are based on the conventional analysis of patterns of cloudiness and involve different empirically defined thresholds of geophysical variables. As a result, various detection methods typically reveal very different results when applied to a single dataset. We develop a conceptually novel approach for the detection of MCs based on the use of deep convolutional neural networks (DCNNs). As a first step, we demonstrate that DCNN model is capable of performing binary classification of 500x500km patches of satellite images regarding MC patterns presence in it. The training dataset is based on the reference database of MCs manually tracked in the Southern Hemisphere from satellite mosaics. We use a subset of this database with MC diameters falling in the range of 200-400 km. This dataset is further used for testing several different DCNN setups, specifically, DCNN built “from scratch”, DCNN based on VGG16 pre-trained weights also engaging the Transfer Learning technique, and DCNN based on VGG16 with Fine Tuning technique. Each of these networks is further applied to both infrared (IR) and a combination of infrared and water vapor (IR+WV) satellite imagery. The best skills (97% in terms of the binary classification accuracy score) is achieved with the model that averages the estimates of the ensemble of different DCNNs. The algorithm can be further extended to the automatic identification and tracking numerical scheme and applied to other atmospheric phenomena characterized by a distinct signature in satellite imagery.

In this article, the evolution in space and in time of the coronavirus pandemic is studied by utilizing a neural network with a self-organizing nature for the spatial analysis of data, and a fuzzy fractal method for capturing the temporal trends of the time series of the countries. Self-organizing neural networks possess the capability for clustering countries in the space domain based on their similar characteristics with respect to their coronavirus cases. In this form enabling finding the countries that are having similar behavior and thus can benefit from utilizing the same methods in fighting the virus propagation. To validate the approach, publicly available datasets of coronavirus cases worldwide have been used. In addition, a fuzzy fractal approach is utilized for the temporal analysis of time series of the countries. Then, a hybrid combination of both the self-organizing maps and the fuzzy fractal approach is proposed for efficient COVID-19 forecasting of the countries. Relevant conclusions have emerged from this study, that may be of great help in putting forward the best possible strategies in fighting the virus pandemic. A lot of the existing works concerned with the Coronavirus have look at the problem mostly from the temporal viewpoint that is of course relevant, but we strongly believe that the combination of both aspects of the problem is relevant to improve the forecasting ability. The most relevant contribution of this article is the proposal of combining neural networks with a self-organizing nature for clustering countries with high similarity and the fuzzy fractal approach for being able to forecast the times series and help in planning control actions for the Coronavirus pandemic.

Deep neural networks (DNN) have successfully delivered a cutting-edge performance in several fields. With the broader deployment of DNN models on critical applications, the security of DNNs becomes an active and yet nascent area. Attacks against DNNs can have catastrophic results, according to recent studies. Poisoning attacks, including backdoor and Trojan attacks, are one of the growing threats against DNNs. Having a wide-angle view of these evolving threats is essential to better understand the security issues. In this regard, creating a semantic model and a knowledge graph for poisoning attacks can reveal the relationships between attacks across intricate data to enhance the security knowledge landscape. In this paper, we propose a DNN Poisoning Attacks Ontology (DNNPAO) that would enhance knowledge sharing and enable further advancements in the field. To do so, we have performed a systematic review of the relevant literature to identify the current state. We collected 28,469 papers from IEEE, ScienceDirect, Web of Science, and Scopus databases, and from these papers, 712 research papers were screened in a rigorous process, and 55 poisoning attacks in DNNs were identified and classified. We extracted a taxonomy of the poisoning attacks as a scheme to develop DNNPAO. Subsequently, we used DNNPAO as a framework to create a knowledge base. Our findings open new lines of research within the field of AI security.

Convolutional Neural Networks combined with autonomous drones are increasingly seen as enablers of partially automating the aircraft maintenance visual inspection process. Such an innovative concept can have a significant impact on aircraft operations. Through supporting aircraft maintenance engineers detect and classify a wide range of defects, the time spent on inspection can significantly be reduced. Examples of defects that can be automatically detected include aircraft dents, paint defects, cracks and holes, and lightning strike damage. Additionally, this concept could also increase the accuracy of damage detection and reduce the number of aircraft inspection incidents related to human factors like fatigue and time pressure. In our previous work, we have applied a recent Convolutional Neural Network architecture known by MASK R-CNN to detect aircraft dents. MASK-RCNN was chosen because it enables the detection of multiple objects in an image while simultaneously generating a segmentation mask for each instance. The previously obtained F1 and F2 scores were 62.67% and 59.35% respectively. This paper extends the previous work by applying different techniques to improve and evaluate prediction performance experimentally. The approaches uses include (1) Balancing the original dataset by adding images without dents; (2) Increasing data homogeneity by focusing on wing images only; (3) Exploring the potential of three augmentation techniques in improving model performance namely flipping, rotating, and blurring; and (4) using a pre-classifier in combination with MASK R-CNN. The results show that a hybrid approache combining MASK R-CNN and augmentation techniques leads to an improved performance with an F1 score of (67.50%) and F2 score of (66.37%)

Applications for Wireless Sensor Networks (WSNs) range from monitoring the environment to automating factories. However, sustained and effective functioning is made more difficult by Sensor Nodes (SNs) limited energy supplies in which optimization is the main issue. So with the aim of increasing the lifespan by decreasing the energy consumption of WSN, Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol with Deep Learning (DL) algorithm is analyzed in this paper. LEACH is a hierarchical mechanism that elects Cluster Heads (CHs) and regularly rotates their positions in order to allocate energy use effectively by using the same amount of energy. However, Deep Learning (DL) method is used to further improve energy optimization. In many applications, the types of Deep Learning methods like Artificial Neural Networks (ANNs) have shown to be very useful. Using this method, WSNs may make more efficient decisions that reduce energy consumption. Data aggregation, duty cycling, and transmission protocols may all be optimized by Deep Learning model's ability to recognize patterns and forecast network behavior. This results in lower energy consumption, a longer lifespan for the network, and better overall performance.

Many engineered approaches have been proposed over the years for solving the hard problem of performing indoor localisation using smartphone sensors. However, specialising these solutions for difficult edge cases remains challenging. Here we propose an end-to-end hybrid multimodal deep neural network localisation system, MM-Loc, relying on zero hand-engineered features, learning them automatically from data instead. This is achieved by using modality-specific neural networks to extract preliminary features from each sensing modality, which are then combined by cross-modality neural structures. We show that our choice of modality-specific neural architectures is capable of estimating the location with good accuracy independently. But for better accuracy, a multimodal neural network fusing the features of early modality-specific representations is a better proposition. Our proposed MM-Loc solution is tested on cross-modality samples characterised by different sampling rates and data representation (inertial sensors, magnetic and WiFi signals), outperforming traditional approaches for location estimation. MM-Loc elegantly trains directly from data unlike conventional indoor positioning systems, which rely on human intuition.

Timely diagnosis of Alzheimer's diseases(AD) is crucial to obtain more practical treatments. In this paper, a novel approach Based on Multi-Level Fuzzy Neural Networks (MLFNN) for early detection of AD is proposed. The focus of study was on the problem of diagnosing AD and MCI patients from healthy people using MLFNN and selecting the best feature(s) and most compatible classification algorithm. In this way, we achieve an excellent performance using only a single feature i.e. MMSE score, by fitting the optimum algorithm to the best area using optimum possible feature(s) namely one feature for a real life problem. It can be said, the proposed method is a discovery that help patients and healthy people get rid of painful and time consuming experiments. Experiments shows the effectiveness of proposed method in current research for diagnosis of AD with one of the highest performance (accuracy rates of 96.6%), ever reported in the literature.

In considering knowledge graphs in a diverse range of domains of interest, graph neural networks have demonstrated significant improvements in node classification and prediction when applied to graph representation with learning node embedding to effectively represent hierarchical properties of graphs. DiffPool is a deep-learning approach using a differentiable graph pooling technique that generates hierarchical representations of graphs. In operation DiffPool is a differentiable graph pooling technique that generates hierarchical representations of graphs. DiffPool learns a differentiable soft cluster assignment for nodes at each layer of a deep graph neural network with nodes mapped sets of clusters. However, control of the learning process is difficult given the complexity and large number of parameters on an `end-to-end’ model. To address this difficulty we propose an novel approach termed FPool which is predicated on the basic approach adopted in DiffPool (where pooling is applied directly to node representations). Methods designed to enhance data classification have been developed and evaluated using a number popular and publicly available sensor data sets. Experimental results for FPool demonstrate improved classification and prediction performance when compared to alternative methods. Moreover, FPool shows an important reduction in the training time over the basic DiffPool framework.