Search

The aim of this research is to predict the mechanical properties along with the behaviors of standard insulating paper used in power transformers under thermal aging. This is conducted by applying an artificial neural network (ANN) trained with a multiple regression model and a particle swarm optimization (MR-PSO) model. The aging of the paper insulation is monitored directly by the tensile strength and the degree of polymerization of the solid insulation and indirectly by chemical markers using 2-furfuraldehyde compound content in oil (2-FAL). A mathematical model is then developed to simulate the mechanical properties (degree of polymerization (DPV) and tensile index (Tidx)) of the aged insulation paper. First, the datasets obtained from experimental results are used to create the MR model, and then the optimizer method PSO is used to optimize its coefficients in order to improve the MR model. Then, an ANN method is trained using the MR-PSO to create a nonlinear correlation between the DPV and the time, temperature, and 2-FAL values. The acquired results are assessed and compared with the experimental data. The model presents almost the same behavior. In particular, it has the capability to accurately simulate the nonlinear property behavior of insulation under thermal aging with an acceptable margin of error. Since the life expectancy of power transformers is directly related to that of the insulating paper, the proposed model can be useful to maintenance planners. [ABSTRACT FROM AUTHOR]

The research groups in computer vision, graphics, and machine learning have dedicated a substantial amount of attention to the areas of 3D object reconstruction, augmentation, and registration. Deep learning is the predominant method used in artificial intelligence for addressing computer vision challenges. However, deep learning on three-dimensional data presents distinct obstacles and is now in its nascent phase. There have been significant advancements in deep learning specifically for three-dimensional data, offering a range of ways to address these issues. This study offers a comprehensive examination of the latest advancements in deep learning methodologies. We examine many benchmark models for the tasks of 3D object registration, augmentation, and reconstruction. We thoroughly analyse their architectures, advantages, and constraints. In summary, this report provides a comprehensive overview of recent advancements in three-dimensional deep learning and highlights unresolved research areas that will need to be addressed in the future. [ABSTRACT FROM AUTHOR]

Forecasting has extreme importance in industry due to the numerous competitive advantages that it provides, allowing to foresee what might happen and adjust management decisions accordingly. Industries increasingly use sensors, which allow for large-scale data collection. Big datasets enable training, testing and application of complex predictive algorithms based on machine learning models. The present paper focuses on predicting values from sensors installed on a pulp paper press, using data collected over three years. The variables analyzed are electric current, pressure, temperature, torque, oil level and velocity. The results of XGBoost and artificial neural networks, with different feature vectors, are compared. They show that it is possible to predict sensor data in the long term and thus predict the asset's behaviour several days in advance. [ABSTRACT FROM AUTHOR]

This special issue contains six revised and extended versions of tool papers that appeared in the proceedings of TACAS 2021, the 27th International Conference on Tools and Algorithms for the Construction and Analysis of Systems. The issue is dedicated to the realization of algorithms in tools and the studies of the application of these tools for analysing hard- and software systems. [ABSTRACT FROM AUTHOR]

Artificial intelligence is more present than ever, both in our society in general and in science. At the center of this development has been the concept of deep learning, the use of artificial neural networks that are many layers deep and can often reproduce human‐like behavior much better than other machine‐learning techniques. The articles in this collection are some recent examples of its application for X‐ray photon science and crystallography that have been published in Journal of Applied Crystallography. [ABSTRACT FROM AUTHOR]

Machine learning (ML), a form of artificial intelligence (AI), is being increasingly employed in neurology. Reported performance metrics often match or exceed the efficiency of average clinicians. The neurologist is easily baffled by the underlying concepts and terminologies associated with ML studies. The superlative performance metrics of ML algorithms often hide the opaque nature of its inner workings. Questions regarding ML model's interpretability and reproducibility of its results in real-world scenarios, need emphasis. Given an abundance of time and information, the expert clinician should be able to deliver comparable predictions to ML models, a useful benchmark while evaluating its performance. Predictive performance metrics of ML models should not be confused with causal inference between its input and output. ML and clinical gestalt should compete in a randomized controlled trial before they can complement each other for screening, triaging, providing second opinions and modifying treatment. [ABSTRACT FROM AUTHOR]

Gait recognition has become an important biometric feature for human identification, in addition to data such as face, iris, and fingerprint. The goal of human gait recognition is to identify people based on walking images. Artificial intelligence technologies have revolutionized the field of gait recognition by enabling computers to automatically learn and extract intricate patterns. These techniques examine video recordings to determine key features in an individual's gait, and these features are used to identify the person. This paper examines the existing appearance-based gait recognition methods that have been published in recent years. The primary objective of this paper is to provide an informative survey of the state-of-the-art in appearance-based gait recognition techniques, highlighting their applications, strengths, and limitations. Through our analysis, we aim to highlight the significant advance that has been made in this field, draw attention to the challenges that have been faced, and identify areas of prospective future research and advances in technology. Furthermore, we comprehensively examine common datasets used in gait recognition research. By analyzing the latest developments in appearance-based gait recognition, our study aims to be a helpful resource for researchers, providing an extensive overview of current methods and guiding future attempts in this dynamic field. [ABSTRACT FROM AUTHOR]

In the modern medical industry, sensors and other medical electronic equipment such as ECG, thermometers, etc. that measure vitals like blood pressure, body temperature, heartbeat, and blood cells are used to record patient data. The information gathered helps to create a patient profile, which in turn helps doctors treat patients appropriately and helps users gain insights into their health. But, the information collected is regionally specific and not worldwide. Thus, this may create problems in transparency, integrity, and much more. The medical details of patients collected are stored in a simple database which appears to be a centralized application that is prone to be modified or hacked by intruders and also suffers from fault tolerance. This paper aims to solve this problem using Blockchain and Data Possession methodologies. Blockchain helps to store multiple copies of data in a decentralized fashion. Thus, a secured model could be implemented using blockchain in which the patient’s data can be securely sent to any medical professional, thus promoting transparency, security, and integrity. Provable Data Possession helps the users to verify that their information is intact and has not been tampered. It can be achieved using cryptography which encrypts the information of the patients stored. This paper will be beneficial for the medical department and the public such that it will be beneficial in terms of transparency, scalability, and security. [ABSTRACT FROM AUTHOR]

Nowadays, landing gears used in aircraft are the basic systems that carry the weight of aircraft systems and provide them to move on the ground. During the landing of aircraft, big and sudden vibrations occur as soon as the wheels come into contact with the runway. This situation negatively affects human safety, comfort and also the life of the landing gear system. Therefore, it is important to keep this vibration at permissible levels. This paper presents an improved vibration control method that controls the amplitudes of vibrations of landing systems. This paper presents proposed vibration control method that controls the angular position variables of the landing gear mechanism to control the vibrations occurring during landing. For different runway conditions and 200 km/h landing speed, the angular joint variables of the landing gear mechanism links are controlled using standard controller and a proposed neural controller structure. The simulation results are shown that the proposed neural controllers have better performance at vibration control of landing systems during landing. [ABSTRACT FROM AUTHOR]

To construct a nonlinear fractional-order neural network reflecting the complex environment of the real world, this paper considers the common factors such as uncertainties, perturbations, and delays that affect the stability of the network system. In particular, not only does the activation function include multiple time delays, but the memristive connection weights also consider transmission delays. Stemming from the characteristics of neural networks, two different types of discontinuous controllers with state information and sign functions are devised to effectuate network synchronization objectives. Combining the finite-time convergence criterion and the theory of fractional-order calculus, Mittag-Leffler synchronization conditions for fractional-order multi-delayed memristive neural networks (FMMNNs) are derived, and the upper bound of the setting time can be confirmed. Unlike previous jobs, this article focuses on applying different inequality techniques in the synchronous analysis process, rather than comparison principles to manage the multi-delay effects. In addition, this study removes the restrictive requirement that the activation function has a zero value at the switching jumps, and the discontinuous control protocol in this paper makes the networks achieve synchronization over a finite time, with some advantages in terms of the convergence speed. [ABSTRACT FROM AUTHOR]

Introduction: Many studies show that words learned early in life are read more easily than the ones learned later and are less vulnerable to brain damage. Methods: the first part of the current study, 25 primary school students in the 5th grade read the word groups learned initially during a previous grade. The words used in the experiments were 327 Farsi monosyllable words matched on the other factors involved in Farsi word naming. Results: The analysis of covariance (the consistency and frequency as covariates) showed that words learned in earlier grades were read more easily than the ones learned later, showing the known effect of the Age of Acquisition (AoA). In the second part of the study, it was tried to simulate AoA in word naming by a neural network model developed earlier based on connectionist approach. While previous studies used random patterns, in the current study words from primary school books were used. Likewise, words learned early by the model were read better than words learned later. However, there was a failure in replicating previous simulation of AoA in English reading by an algorithm called Quick prop for Farsi. In addition, the model was lesioned by removing some hidden units to see its effect on word reading. As a result, words learned earlier were less vulnerable to damage compared with the ones learned later. Conclusion: The findings showed that words learned earlier, compared to those learned later, were read better and were less vulnerable to damage. These effects are explained by considering the nature of learning in neural networks trained by error back-propagation. [ABSTRACT FROM AUTHOR]

Thomas L. Saaty, a world-renowned scholar and our beloved Advisory Editor, passed away on August 14, 2017 at the age of 91. He has made many fundamental contributions to operations research, analytics, business and mathematics. Despite his heavy illness for 14 months, as passionate about science, he has also enthusiastically published his research. We review his final works published in 2017 and 2018. We are also honored to reveal and publish his final statements on neural firing and synthesis in making comparisons & life satisfaction, respectively, in his final two papers; the previous one submitted by him just before his passing and the other one submitted by his co-authors after his death to European Journal of Pure and Applied Mathematics (EJPAM). We just would like to remember and honor Saaty's memory by publishing his final papers at EJPAM and this humble remarks dedicated to the memory of Thomas L. Saaty on the occasion of the first anniversary of his passing, August 14, 2018. [ABSTRACT FROM AUTHOR]

This paper considers the problem of target tracking in complex manoeuvering scenarios with a lack of relevant prior knowledge. This is a challenge for classical model‐based manoeuvering target tracking algorithms because they rely heavily on accurate domain and prior knowledge of target motion. To address this problem, we propose a hybrid‐driven continuous‐time filter algorithm in this paper, which combines the advantages of the model‐driven and data‐driven. We use the stochastic differential equation (SDE) with the acceleration model as the basic framework of the proposed algorithm. In order to deal with unpredictable manoeuvres and unknown perturbations, we adopt neural networks as data‐driven to estimate target accelerations and compensations of composite perturbations in real time using historical and current measurements. And the direction of gradient descent of the neural network is constrained by the motion model, thus the learning efficiency of the network and the interpretability of the proposed algorithm is improved. The proposed algorithm can simultaneously utilise historical trajectory information and domain knowledge as hybrid‐driven to achieve complex manoeuvering target tracking with little prior information. Experimental results demonstrate the superiority of our algorithm in tracking accuracy, robustness and environmental adaptability. [ABSTRACT FROM AUTHOR]

De-mining operations are of critical importance for humanitarian efforts and safety in conflict-affected regions. In this paper, we address the challenge of enhancing the accuracy and efficiency of mine detection systems. We present an innovative Deep Learning architecture tailored for pulse induction-based Metallic Mine Detectors (MMD), so called DL-MMD. Our methodology leverages deep neural networks to distinguish amongst nine distinct materials with an exceptional validation accuracy of 93.5%. This high level of precision enables us not only to differentiate between anti-personnel mines, without metal plates but also to detect minuscule 0.2-g vertical paper pins in both mineralized soil and non-mineralized environments. Moreover, through comparative analysis, we demonstrate a substantial 3% and 7% improvement (approx.) in accuracy performance compared to the traditional K-Nearest Neighbors and Support Vector Machine classifiers, respectively. The fusion of deep neural networks with the pulse induction-based MMD not only presents a cost-effective solution but also significantly expedites decision-making processes in de-mining operations, ultimately contributing to improved safety and effectiveness in these critical endeavors. [ABSTRACT FROM AUTHOR]

Black heart disease is one of the screening indicators of seed potatoes, which has a serious impact on the quality and yield of potato, and at present there are fewer non-destructive testing methods for internal defects of seed potatoes. This paper aims to utilize non-destructive testing technology to quickly identify qualified and black hearted seed potatoes, and then to protect yield and food security. In this paper, transmission spectroscopy system was utilized to collect the spectral data of 104 qualified seed potatoes and 104 black hearted seed potatoes in 450~940 nm band. Subsequently, four algorithms, namely Savitzky-Golay (SG), Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC) and First-order Derivative (FD), were utilized to pre-process the seed potatoes spectral data to improve the data noise ratio. Feature wavelength extraction was made using Competitive Adaptive Reweighted Sampling (CARS) and Successive Projections Algorithm (SPA) to enhance the sample data characteristics and improve the model interpretability. The construction of classification models for qualified and black hearted seed potatoes relied on two deep learning techniques, Convolutional Neural Networks (CNN) and Recurrent Neural Network (RNN), which were trained and tested for the feature bands respectively. The experimental results showed that SG-CARS-CNN was the optimal combination of classification algorithms, and the classification accuracies of both the training set and the test set reached 100%, which improved the accuracy of the test set by 3.85% compared with that of the traditional machine learning algorithms, and provided an accurate method for the rapid screening of qualified seed potatoes. [ABSTRACT FROM AUTHOR]

Image semantic segmentation technology is one of the core research contents in the field of computer vision. With the improvement of computer performance and the continuous development of deep learning technology, researchers have more and more enthusiasm to study the actual effect and performance of image semantic segmentation. The results of deep semantic segmentation allow computers to have a more detailed and accurate understanding of images, and have a wide range of application needs in the fields of autonomous driving, intelligent security, medical imaging, remote sensing images, etc. However, the existing image semantic segmentation algorithms have the disadvantages of easy discontinuous results and insufficient prediction accuracy. In this paper, we take deep learning-based image semantic segmentation technology as the research object to explore the improvement of the image semantic segmentation algorithm and its application in road scenarios. First, this paper proposes MCU-Net method based on residual fusion and multi-scale contextual information. MCU-Net uses residual fusion module to deepen the network structure and improve the ability of U-Net to acquire deeper features. Then a top-down and bottom-up path is constructed for feature information between different levels, and the spatial and semantic information contained in shallow and deep features in the network is fully utilized by fusing features from different levels. In addition, an enhanced void space pyramid pooling module is added for feature information between the same levels, which enables the output features to have a larger range of semantic information. Second, this paper proposes the DAMCU-Net method based on attention mechanism and edge detection based on MCU-Net. DAMCU-Net extracts global contextual information by the attention mechanism optimization module, while fusing features using dense jump connections to facilitate the network to recover more spatial detail information during upsampling, and uses the FReLU activation function to improve the segmentation capability of the network for complex targets. For the edge information lost in the feature extraction process, the edge detection branch is added to supplement the feature information of the main path by feature fusion to achieve the optimization of the edge information. [ABSTRACT FROM AUTHOR]

In this paper, we present a sharper version of the results in the paper [H. N. Mhaskar, Dimension independent bounds for general shallow networks, Neural Netw. 123 (2020) 142–152]. Let and be compact metric spaces. We consider approximation of functions of the form x ↦ ∫ G (x , y) d τ (y) , x ∈ , by G -networks of the form x ↦ ∑ k = 1 n a k G (x , y k) , y 1 , ... , y n ∈ , a 1 , ... , a n ∈ ℝ. Defining the dimensions of and in terms of covering numbers, we obtain dimension independent bounds on the degree of approximation in terms of n , where also the constants involved are all dependent at most polynomially on the dimensions. Applications include approximation by power rectified linear unit networks, zonal function networks, certain radial basis function networks as well as the important problem of function extension to higher-dimensional spaces. [ABSTRACT FROM AUTHOR]

To achieve high precision and robust control of switched reluctance machines (SRMs), this paper proposes a novel rotor position estimation method based on gain-optimized sliding mode observer (SMO) with neural networks and whale optimization algorithm. For the SMO, this paper uses the easily measured phase current to constitute the error function of the SMO, which requires less pre-stored data. Using the neural networks' powerful nonlinear mapping capability, the relationship between the sliding mode gains and speed estimation error and the position estimation error is obtained, and then the whale optimization algorithm is used to find the optimal sliding mode gains in the restricted range. The simulation and experiment results show that the accuracy of the SMO control system is significantly improved. [ABSTRACT FROM AUTHOR]

Papermaking is an industrial process that is becoming more competitive nowadays. In this process there are numerous techniques and measurements to indicate paper quality. To increase competitiveness a good control of paper quality is needed through paper properties predictions from different process measurements. However, complex physico-chemical processes take place during papermaking, and thus, paper property predictions are not easy to obtain, especially in the wet-end area. In the wet end flocculation takes place, which will determine the floc properties during the formation of the sheet, and therefore, it will influence retention, drainage and formation. These strongly affect the runnability of the machine and the properties of the final product, and thus, using wet-end measurements for the predictions implies advanced data treatment. Artificial neural networks have been used in this article to predict newsprint paper properties from wet-end parameters. Results show that formation and strength properties can be robustly predicted from pulp properties at the headbox, flocculation parameters and machine speed. [ABSTRACT FROM AUTHOR]

This survey introduces 101 new publications on applications of Clifford's geometric algebras (GAs) newly published during 2022 (until mid‐January 2023). The selection of papers is based on a comprehensive search with Dimensions.ai, followed by detailed screening and clustering. Readers will learn about the use of GA for mathematics, computation, surface representations, geometry, image, and signal processing, computing and software, quantum computing, data processing, neural networks, medical science, physics, electric engineering, control and robotics. [ABSTRACT FROM AUTHOR]

Organic‐chemical literature encompasses large numbers of catalysts and reactions they can effect. Many of these examples are published merely to document the catalysts' scope but do not necessarily guarantee that a given catalyst is "optimal"—in terms of yield or enantiomeric excess—for a particular reaction. This paper describes a Machine Learning model that aims to improve such catalyst‐reaction assignments based on the carefully curated literature data. As we show here for the case of asymmetric magnesium catalysis, this model achieves relatively high accuracy and offers out of‐the‐box predictions successfully validated by experiment, e.g. in synthetically demanding asymmetric reductions or Michael additions. [ABSTRACT FROM AUTHOR]

Organic‐chemical literature encompasses large numbers of catalysts and reactions they can effect. Many of these examples are published merely to document the catalysts' scope but do not necessarily guarantee that a given catalyst is "optimal"—in terms of yield or enantiomeric excess—for a particular reaction. This paper describes a Machine Learning model that aims to improve such catalyst‐reaction assignments based on the carefully curated literature data. As we show here for the case of asymmetric magnesium catalysis, this model achieves relatively high accuracy and offers out of‐the‐box predictions successfully validated by experiment, e.g. in synthetically demanding asymmetric reductions or Michael additions. [ABSTRACT FROM AUTHOR]

In this paper, we develop an integral reinforcement learning (IRL)‐based trajectory tracking control (TTC) scheme via firefly optimized neural networks for partially unknown multiplayer nonlinear systems. Under the developed TTC scheme, IRL is proved to be equivalent to the classical policy iteration, which guarantees the convergence of the IRL algorithm. By implementing the IRL method, the requirement of the drift dynamics is obviated. The TTC policy for each player is obtained by solving the coupled Hamilton–Jacobi equation with a critic neural network, whose weight vector is tuned by the firefly algorithm. The tracking error of the closed‐loop system is guaranteed to be stable via the Lyapunov's direct method. Simulation results illustrate the effectiveness and superiority of the present IRL‐based TTC scheme, and show that the success rate of system operation is increased by introducing the firefly algorithm. [ABSTRACT FROM AUTHOR]

This paper investigates the practical prescribed time tracking for a class of uncertain nonlinear systems based on neural networks and event‐triggered control. Introducing a time‐varying constraint function transforms the original practical prescribed time‐tracking control issue into a tracking error constraint problem. An event‐triggered adaptive control has been proposed, which can effectively reduce the communication burden between the controller and the actuator. Using neural networks to approximate unknown nonlinear functions avoids the differentiation of virtual controllers, thereby reducing the computational burden. In addition, users can independently choose preset time and tracking accuracy without changing the control structure, which remains independent of the initial conditions and any design parameters. Finally, the effectiveness of this method is verified through simulation examples. [ABSTRACT FROM AUTHOR]

In the production process of raw liquid colored polyester, there will be color changes from fibers to yarns and from yarns to fabrics. The color changes of yarns and fabrics are often not caused by the color changes of the fibers themselves. There are many factors that can cause color changes, such as yam thickness, twist, twist direction, and fabric structure. At present, the judgment of color changes from fibers to yarns and fabrics by enterprises mainly relies on manual experience, which is subjective and difficult to control. Therefore, accurately predicting the color patterns between fibers, yarns, and fabrics has practical application value for enterprise production. In recent years, many domestic researchers have developed many theoretical models for color prediction in the textile field. Traditional color matching models include the Friele model, Stearns-Noechel model, and Kubelka-Munk model, which have certain limitations. To control the regularity of color parameters L, a, and b in the production process of polyester fibers, yarns, and fabrics dyed with raw liquid, and to improve the accuracy of color prediction between fibers, yarns, and fabrics, this paper took the samples of polyester fibers, yarns, and fabrics dyed with raw liquids produced by enterprises as the research object and proposed a color prediction method based on neural networks, which are intelligent computing methods that simulate biological neural networks in computer network systems. Firstly, a colorimeter was used to obtain the L, a, and b values of 288 sets of color samples, each containing fiber, yarn, and fabric samples of the same color. Then, the samples were divided into three training groups, namely fiber yarn group, yarn fabric group, and fiber fabric group. Simultaneously, the data were input into the network for modeling, with fiber groups as inputs and yarn groups as targets; the yarn group served as the input, and the fabric group served as the target; fibers served as input and fabrics as target. Finally, the performance of the neural network was adjusted based on the average Eus of the network and the color difference between the output and the true color. The training algorithm, number of neurons, and transfer function were adjusted separately. This article expanded the sample data to 1, 000 groups by adding noise, enhancing the network's generalization ability and improving the accuracy of experimental results. The experimental results show that with trainlm as the training algorithm, when the number of neurons is 100 and the transfer functions of the network input and hidden layer and hidden layer and output are tansig and purelin, respectively, the EMS of the network is below 1.20, and the color difference is between 0.52 and 0. 64, the network has good predictive performance. The average color difference of the test group using the trained network for coloring polyester staple fibers, yarns, and fabrics in the original solution is less than 0.7. indicating that the network training effect is good. The use of neural networks for color prediction presents a new method for color prediction, and the results of this study can provide reference for enterprises to control color changes in textile production processes. [ABSTRACT FROM AUTHOR]

Many authors have been examined simultaneous approximation by neural networks with one hidden layer. Hahm and Hong suggested some simultaneous approximation results for univariate functions. Hahm and Hong obtained degree of multivariate functions by generalized translation networks with one hidden layer. In this paper, we suggest degree of simultaneous approximation of multivariate function by neural networks with one hidden layer. [ABSTRACT FROM AUTHOR]

Authentication is critical for Internet-of-Things. The traditional approach of using cryptographic keys is subject to invasive attacks. Being unclonable even by the manufacturers, physical unclonable functions (PUFs) leverage integrated circuits' manufacturing variations to produce responses unique for individual devices, and hence are of great potential as security primitives. While physically unclonable, many PUFs were reported to be mathematically cloneable by machine learning-based modeling methods. The feed-forward arbiter PUFs (FF PUFs) are among the PUFs with strong resistance against machine learning attacks. Existing studies revealed that only a very small group of FF PUFs with special loop patterns had been broken, and the vast majority of FF PUFs are still secure against all machine learning attack methods tried so far. In this paper, we introduce a neural network that can successfully attack FF PUFs with any loop patterns, with training time even magnitudes lower than existing methods attacking PUFs of the restrictive loop patterns. Experimental results show that, on the one hand, FF PUFs are not secure against attacks even with a large number of complex feed-forward loops, hence susceptible to attacks by response-prediction-based malicious software. On the other hand, the new approach of designing problem-tailored attack methods points to a new way to identify PUF security risks which might be difficult to discover by general-purpose machine learning methods. [ABSTRACT FROM AUTHOR]

Forward–backward doubly stochastic differential equations (FBDSDEs) serve as a probabilistic interpretation of stochastic partial differential equations (SPDEs) with diverse applications. Coupled FBDSDEs encounter numerous challenges in numerical approximation compared to forward–backward stochastic differential equations (FBSDEs) and decoupled FBDSDEs, including ensuring the measurability of the numerical solutions, accounting for the mutual influences between forward and backward processes, and considering the relationship with respect to SPDEs rather than PDEs. This paper introduces, for the first time, a numerical method for solving multi-dimensional coupled FBDSDEs. By integrating an optimal control-based approach with deep neural networks, it effectively addresses the coupling-related challenges between forward and backward equations. Computational examples of coupled FBDSDEs with explicit solutions demonstrate that the proposed deep learning-based numerical algorithm achieves commendable performance in terms of both accuracy and efficiency. [ABSTRACT FROM AUTHOR]

Accurate numerical values of aerodynamic parameters are important in aircraft design. The knowledge of stability and control aerodynamic parameters is essential to postulate high-fidelity control laws. The aerodynamic forces and moments are strong functions of the angle of attack (AOA), Reynolds number, and control surface deflections. Typically, conventional estimation techniques such as maximum likelihood (MLE) and least-squares (LS) principles facilitate the determination of these parameters. Unsteady aerodynamics may complicate the estimation of aerodynamic parameters at high AOA. Data-driven techniques employing neural networks provide an alternative for modeling the system behavior based on its observed state and control input variables. Nonlinearity increases because of flow separation at high AOA, which is close to stall. This paper explores the feasibility of employing a machine learning approach using neural networks to predict aircraft dynamics in a limited sense to identify aerodynamic characteristics. Integrating a neural network with the artificial bee colony (ABC) method facilitated the optimization of unknowns of the proposed aerodynamic model (AM). The proposed neural artificial bee colony (NABC) optimization approach estimated the longitudinal dynamics and stall properties for two experimental aircraft. Comparison of the estimates provided by the NABC approach with those of the standard MLE and neural Gauss–Newton (NGN) techniques established its efficacy. Furthermore, robust statistical analysis indicated that the proposed method provides a viable alternative for parameter estimation in nonlinear applications. [ABSTRACT FROM AUTHOR]

Recently, accurate forecasting of mortality rates with deep learning models has been investigated in several papers in the actuarial literature. Most of the models proposed to date are not explainable, making it difficult to communicate the basis on which mortality forecasts have been made. We adapt the LocalGLMnet of Richman, R. & Wüthrich, M. V. (2023). [LocalGLMnet: Interpretable deep learning for tabular data. Scandinavian Actuarial Journal2023(1), 71–95] to produce explainable forecasts of mortality rates using locally connected neural networks, and we show that these can be interpreted as autoregressive time-series models of mortality rates. These forecasts are shown to be highly accurate on the Human Mortality Database and the United States Mortality Database. Finally, we show how regularizing the LocalGLMnet can produce improved forecasts, and that by applying auto-encoders, observations of mortality rates can be denoised to improve forecasts even further. [ABSTRACT FROM AUTHOR]

The primary focus of this research paper is to explore the realm of dynamic learning in sampled‐data strict‐feedback nonlinear systems (SFNSs) by leveraging the capabilities of radial basis function (RBF) neural networks (NNs) under the framework of adaptive control. First, the exact discrete‐time model of the continuous‐time system is expressed as an Euler strict‐feedback model with a sampling approximation error. We provide the consistency condition that establishes the relationship between the exact model and the Euler model with meticulous detail. Meanwhile, a novel lemma is derived to show the stability condition of a digital first‐order filter. To address the non‐causality issues of SFNSs with sampling approximation error and the input data dimension explosion of NNs, the auxiliary digital first‐order filter and backstepping technology are combined to propose an adaptive neural dynamic surface control (ANDSC) scheme. Such a scheme avoids the n$$ n $$‐step time delays associated with the existing NN updating laws derived by the common n$$ n $$‐step predictor technology. A rigorous recursion method is employed to provide a comprehensive verification of the stability, guaranteeing its overall performance and dependability. Following that, the NN weight error systems are systematically decomposed into a sequence of linear time‐varying subsystems, allowing for a more detailed analysis and understanding. In order to ensure the recurrent nature of the input variables, a recursive design is employed, thereby satisfying the partial persistent excitation condition specifically designed for the RBF NNs. Meanwhile, it can verify that the NN estimated weights converge to their ideal values. Compared with the common n$$ n $$‐step predictor technology, there is no need to redesign the learning rules due to the designed NN weight updating laws without time delays. Subsequently, after capturing and storing the convergence weights, a novel neural learning dynamic surface control (NLDSC) scheme is specifically formulated by leveraging the acquired knowledge. The introduced methodology reduces computational complexity and facilitates practical implementation. Finally, empirical evidence obtained from simulation experiments validates the efficacy and viability of the proposed methodology. [ABSTRACT FROM AUTHOR]

Autoencoders are dimension reduction models in the field of machine learning which can be thought of as a neural network counterpart of principal components analysis (PCA). Due to their flexibility and good performance, autoencoders have been recently used for estimating nonlinear factor models in finance. The main weakness of autoencoders is that the results are less explainable than those obtained with the PCA. In this paper, we propose the adoption of the Shapley value to improve the explainability of autoencoders in the context of nonlinear factor models. In particular, we measure the relevance of nonlinear latent factors using a forecast-based Shapley value approach that measures each latent factor's contributions in determining the out-of-sample accuracy in factor-augmented models. Considering the interesting empirical instance of the commodity market, we identify the most relevant latent factors for each commodity based on their out-of-sample forecasting ability. [ABSTRACT FROM AUTHOR]

The incorporation of conformal cooling systems has significantly enhanced the efficiency and quality of injection molding process. While several automated methods have been developed for creating conformal cooling channels in injection molds, the current optimization process for conformal cooling design parameters is hindered by labor-intensive iterative thermal simulation processes and the substantial reliance on empirical human knowledge. This paper presents an innovative machine learning method to assess the thermal performance of conformal cooling systems by employing a combination of a non-linear regression model and a neural network. By employing a logarithmic regression model describing the temperature graph and a neural network predicting the coefficients of the logarithmic regression model, the thermal performance of specified conformal cooling systems can be assessed and predicted precisely. This methodology empowers designers to evaluate the thermal efficiency of conformal cooling systems efficiently and effectively to further optimize the conformal cooling design parameters without relying on tedious manual thermal and fluid simulation processes. [ABSTRACT FROM AUTHOR]

In cybersecurity, live production data for predictive analysis pose a significant challenge due to the inherently secure nature of the domain. Although there are publicly available, synthesized, and artificially generated datasets, authentic scenarios are rarely encountered. For anomaly-based detection, the dynamic definition of thresholds has gained importance and attention in detecting abnormalities and preventing malicious activities. Unlike conventional threshold-based methods, deep learning data modeling provides a more nuanced perspective on network monitoring. This enables security systems to continually refine and adapt to the evolving situation in streaming data online, which is also our goal. Furthermore, our work in this paper contributes significantly to AIOps research, particularly through the deployment of our intelligent module that cooperates within a monitoring system in production. Our work addresses a crucial gap in the security research landscape toward more practical and effective secure strategies. [ABSTRACT FROM AUTHOR]

This paper presents a new edge detection process implemented in an embedded IoT device called Bee Smart Detection node to detect catastrophic apiary events. Such events include swarming, queen loss, and the detection of Colony Collapse Disorder (CCD) conditions. Two deep learning sub-processes are used for this purpose. The first uses a fuzzy multi-layered neural network of variable depths called fuzzy-stranded-NN to detect CCD conditions based on temperature and humidity measurements inside the beehive. The second utilizes a deep learning CNN model to detect swarming and queen loss cases based on sound recordings. The proposed processes have been implemented into autonomous Bee Smart Detection IoT devices that transmit their measurements and the detection results to the cloud over Wi-Fi. The BeeSD devices have been tested for easy-to-use functionality, autonomous operation, deep learning model inference accuracy, and inference execution speeds. The author presents the experimental results of the fuzzy-stranded-NN model for detecting critical conditions and deep learning CNN models for detecting swarming and queen loss. From the presented experimental results, the stranded-NN achieved accuracy results up to 95%, while the ResNet-50 model presented accuracy results up to 99% for detecting swarming or queen loss events. The ResNet-18 model is also the fastest inference speed replacement of the ResNet-50 model, achieving up to 93% accuracy results. Finally, cross-comparison of the deep learning models with machine learning ones shows that deep learning models can provide at least 3–5% better accuracy results. [ABSTRACT FROM AUTHOR]

In this paper, the experimental results on microphone virtualization in realistic automotive scenarios are presented. A Temporal Convolutional Network (TCN) was designed in order to estimate the acoustic signal at the driver's ear positions based on the knowledge of monitoring microphone signals at different positions—a technique known as virtual microphone. An experimental setup was implemented on a popular B-segment car to acquire the acoustic field within the cabin while running on smooth asphalt at variable speeds. In order to test the potentiality of the TCN, microphone signals were recorded in two different scenarios, either with or without the front passenger. Our experimental results show that, when training is performed in both scenarios, the adopted TCN is able to robustly adapt to different conditions and guarantee a good average performance. Furthermore, an investigation on the parameters of the Neural Network (NN) that guarantee the sufficient accuracy of the estimation of the virtual microphone signals while maintaining a low computational complexity is presented. [ABSTRACT FROM AUTHOR]

Domain generalization (DG) aims to learn knowledge from multiple related domains to achieve a robust generalization performance in unseen target domains, which is an effective approach to mitigate domain shift in remote sensing image classification. Although the sharpness-aware minimization (SAM) method enhances DG capability and improves remote sensing image classification performance by promoting the convergence of the loss minimum to a flatter loss surface, the perturbation loss (maximum loss within the neighborhood of a local minimum) of SAM fails to accurately measure the true sharpness of the loss landscape. Furthermore, its variants often overlook gradient conflicts, thereby limiting further improvement in DG performance. In this paper, we introduce implicit sharpness-aware minimization (ISAM), a novel method that addresses the deficiencies of SAM and mitigates gradient conflicts. Specifically, we demonstrate that the discrepancy in training loss during gradient ascent or descent serves as an equivalent measure of the dominant eigenvalue of the Hessian matrix. This discrepancy provides a reliable measure for sharpness. ISAM effectively reduces sharpness and mitigates potential conflicts between gradients by implicitly minimizing the discrepancy between training losses while ensuring a sufficiently low minimum through minimizing perturbation loss. Extensive experiments and analyses demonstrate that ISAM significantly enhances the model's generalization ability on remote sensing and DG datasets, outperforming existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Speech Emotion Recognition (SER) is a challenging task, and the typical convolutional neural network (CNN) cannot well handle the speech data directly. Because CNN tends to understand local information and ignores the overall characteristics. This paper proposes a Capsule Network with Two-Way Attention MechanismTWACapsNet for short) for the SER problem. TWACapsNet accepts the spatial and spectral features as inputs, and the convolutional layer and the capsule layer are deployed to process these two types of features in two ways separately. After that, two attention mechanisms are designed to enhance the information obtained from the spatial and spectral features. Finally, the results of these two ways are combined to form the final decision. The advantage of TWACapsNet is verified by experiments on multiple SER data sets, and experimental results show that the proposed method outperforms the widely-deployed neural network models on three typical SER data sets. Furthermore, the combination of the two ways contributes to the higher and more stable performance of TWACapsNet. [ABSTRACT FROM AUTHOR]

Reports produced by popular malware analysis services showed a disparity in samples available for different malware families. The unequal distribution between such classes can be attributed to several factors, such as technological advances and the application domain that seeks to infect a computer virus. Recent studies have demonstrated the effectiveness of deep learning (DL) algorithms when learning multi-class classification tasks using imbalanced datasets. This can be achieved by updating the learning function such that correct and incorrect predictions performed on the minority class are more rewarded or penalized, respectively. This procedure can be logically implemented by leveraging the deep reinforcement learning (DRL) paradigm through a proper formulation of the Markov decision process (MDP). This paper proposes SINNER, i.e., a DRL-based multi-class classifier that approaches the data imbalance problem at the algorithmic level by exploiting a redesigned reward function, which modifies the traditional MDP model used to learn this task. Based on the experimental results, the proposed formula appears to be successful. In addition, SINNER has been compared to several DL-based models that can handle class skew without relying on data-level techniques. Using three out of four datasets sourced from the existing literature, the proposed model achieved state-of-the-art classification performance. [ABSTRACT FROM AUTHOR]

In this paper, the problem of sampled-data adaptive tracking control for a class of switched nonlinear systems with prescribed performance is considered. In order to guarantee the system is stable and achieves the prescribed performance in sampled-data control, a coordinate transformation satisfying the prescribed performance is introduced. In addition, the neural networks (NNs) used to approximate the unknown nonlinear functions and the backstepping technique are applied to design the sampled-data controller and the adaptive laws. An upper bound on the sampling period is obtained to maintain the stability of the systems. It is confirmed that the designed sampled-data scheme ensures that all signals of the closed-loop system are semi-globally uniformly ultimately bounded (SGUUB) and the tracking error is limited to the prescribed performance function. The effectiveness of the designed control scheme is demonstrated by two simulation examples. [ABSTRACT FROM AUTHOR]

The complete characterization of the almost-entropic region yields rate regions for network coding problems. However, this characterization is difficult and open. In this paper, we propose a novel algorithm to determine whether an arbitrary vector in the entropy space is entropic or not, by parameterizing and generating probability mass functions by neural networks. Given a target vector, the algorithm minimizes the normalized distance between the target vector and the generated entropic vector by training the neural network. The algorithm reveals the entropic nature of the target vector, and obtains the underlying distribution, accordingly. The proposed algorithm was further implemented with convolutional neural networks, which naturally fit the structure of joint probability mass functions, and accelerate the algorithm with GPUs. Empirical results demonstrate improved normalized distances and convergence performances compared with prior works. We also conducted optimizations of the Ingleton score and Ingleton violation index, where a new lower bound of the Ingleton violation index was obtained. An inner bound of the almost-entropic region with four random variables was constructed with the proposed method, presenting the current best inner bound measured by the volume ratio. The potential of a computer-aided approach to construct achievable schemes for network coding problems using the proposed method is discussed. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNN) have fundamentally revolutionized the artificial intelligence (AI) field. The transformer model is a type of DNN that was originally used for the natural language processing tasks and has since gained more and more attention for processing various kinds of sequential data, including biological sequences and structured electronic health records. Along with this development, transformer-based models such as BioBERT, MedBERT, and MassGenie have been trained and deployed by researchers to answer various scientific questions originating in the biomedical domain. In this paper, we review the development and application of transformer models for analyzing various biomedical-related datasets such as biomedical textual data, protein sequences, medical structured-longitudinal data, and biomedical images as well as graphs. Also, we look at explainable AI strategies that help to comprehend the predictions of transformer-based models. Finally, we discuss the limitations and challenges of current models, and point out emerging novel research directions. [ABSTRACT FROM AUTHOR]

The goal of the growing discipline of neuro-symbolic artificial intelligence (AI) is to develop AI systems with more human-like reasoning capabilities by combining symbolic reasoning with connectionist learning. We survey the literature on neuro-symbolic AI during the last two decades, including books, monographs, review papers, contribution pieces, opinion articles, foundational workshops/talks, and related PhD theses. Four main features of neuro-symbolic AI are discussed, including representation, learning, reasoning, and decision-making. Finally, we discuss the many applications of neuro-symbolic AI, including question answering, robotics, computer vision, healthcare, and more. Scalability, explainability, and ethical considerations are also covered, as well as other difficulties and limits of neuro-symbolic AI. This study summarizes the current state of the art in neuro-symbolic artificial intelligence. [ABSTRACT FROM AUTHOR]

Generative adversarial networks (or GANs) are a specific deep learning architecture often used for different usages, such as data generation or image-to-image translation. In recent years, this structure has gained increased popularity and has been used in different fields. One area of expertise currently in vogue is the use of GANs to produce synthetic aperture radar (SAR) data, and especially expand training datasets for SAR automatic target recognition (ATR). In effect, the complex SAR image formation makes these kind of data rich in information, leading to the use of deep networks in deep learning-based methods. Yet, deep networks also require sufficient data for training. However, contrary to optical images, we generally do not have a substantial number of available SAR images because of their acquisition and labelling cost; GANs are then an interesting tool. Concurrently, how to improve explainability for SAR ATR deep neural networks and how to make their reasoning more transparent have been increasingly explored as model opacity deteriorates trust of users. This paper aims at reviewing how GANs are used with SAR images, but also giving perspectives on how GANs could be used to improve interpretability and explainability of SAR classifiers. [ABSTRACT FROM AUTHOR]

CFRP hybrid bonded–bolted (HBB) joints combine the advantages of traditional joining methods, namely adhesive bonding, and bolting, to achieve optimal connection performance, making them the most favored connection method. The structural parameters of CFRP HBB joints, including overlap length, bolt-hole spacing, and fit clearance relationships, have a complex impact on connection performance. To enhance the connectivity performance of joint structures, this paper develops a multiscale finite element analysis model to investigate the impact of structural parameters on the strength of CFRP HBB joint structures. Coupled with experimental validation, the study reveals how changes in structural parameters affect the unidirectional tensile failure force of the joints. Building on this, an analytical approach and inverse design methodology for the mechanical properties of CFRP HBB joints based on deep supervised learning algorithms are developed. Neural networks accurately and efficiently predict the performance of joints with unprecedented combinations of parameters, thus expediting the inverse design process. This research combines experimentation and multiscale finite element analysis to explore the unknown relationships between the mechanical properties of CFRP HBB joints and their structural parameters. Furthermore, leveraging DNN neural networks, a rapid calculation method for the mechanical properties of hybrid joints is proposed. The findings lay the groundwork for the broader application and more intricate design of composite materials and their connection structures. [ABSTRACT FROM AUTHOR]

Spotting a significant number of drones flying near the entrance of a beehive during late Spring could indicate the occurrence of swarming mood, as the the surge in drone presence is related to an overcrowded hive. Swarming refers to a natural reproductive process witnessed in honey bees, wherein half of the bee colony departs from their hive alongside the aging queen. In the paper, we propose an early swarming detection mechanism that relies on the behavior of the drones. The proposed method is based on audio signals registered in a close proximity to the beehive entrance. A comparative study was performed to find the most effective preprocessing method for the audio signals for the detection problem. We have compared the results for three different power spectrum density coefficients estimation methods, which are used as an input of an autoencoder neural network to discriminate drones from worker bees. Through simulations employing real-life signals, it has been demonstrated that drone detection based solely on audio signals is indeed feasible. The attained level of detection accuracy enables the creation of an efficient alarm system for beekeepers. [ABSTRACT FROM AUTHOR]

Applying parallel-in-time algorithms to multiscale Hamiltonian systems to obtain stable long-time simulations is very challenging. In this paper, we present novel data-driven methods aimed at improving the standard parareal algorithm developed by Lions et al. in 2001, for multiscale Hamiltonian systems. The first method involves constructing a correction operator to improve a given inaccurate coarse solver through solving a Procrustes problem using data collected online along parareal trajectories. The second method involves constructing an efficient, high-fidelity solver by a neural network trained with offline generated data. For the second method, we address the issues of effective data generation and proper loss function design based on the Hamiltonian function. We show proof-of-concept by applying the proposed methods to a Fermi-Pasta-Ulam (FPU) problem. The numerical results demonstrate that the Procrustes parareal method is able to produce solutions that are more stable in energy compared to the standard parareal. The neural network solver can achieve comparable or better runtime performance compared to numerical solvers of similar accuracy. When combined with the standard parareal algorithm, the improved neural network solutions are slightly more stable in energy than the improved numerical coarse solutions. [ABSTRACT FROM AUTHOR]

The management of large enterprises influences their efficiency and profitability. One of the important aspects is the appropriate management of electricity consumption used for production and daily operation. The problem becomes more complicated when you need to manage not one but a large complex of buildings with heterogeneous purposes. In the paper, we examine real-time series data of electrical energy consumption in a complex of heterogeneous buildings, including offices and warehouses, using time series analysis methods such as the Holt–Winters model and ARIMA/SARIMA model, and neural networks (Deep Neural Network, Recurrent Neural Network, and Long Short-Term Memory). Experimental research was performed on a dataset obtained from an energy consumption meter placed in the building complex, built in different periods, and equipped with a variety of automation devices. The data were collected over a period of four years 2018–2021 in the form of time series. Results show that classic models are good at predicting energy consumption in the mentioned type of buildings. The ARIMA model gave the best results—for buildings characterized by seasonality and trends the forecasts were almost perfect with actual values. [ABSTRACT FROM AUTHOR]