Search

Numerous Artificial Intelligence (AI) solutions are available for achieving thermal comfort. They were either trained with limited datasets or using personalized training with limited field studies. This work assessed the model that used the ASHRAE multiple databases as the shallow supervised learning dataset for an Artificial Neural Network (ANN) based controller suitable for the residential dwellings' node. The learning accuracy can be increased to 96.1%. This paper presented the field studies to show the model performances for the common UK dwellings: the prior 1970s, the new, modular, refurbished, and the use of new materials to improve indoor thermal performance. The result shows that the model was able to perform in different environments and able to acknowledge adaptive human comfort. This was shown by the ability to represent 98.90% of the ASHRAE Standard 55 data, 6.06% improvement from the previous research. As a result, the broader comfort zone acknowledgement can lead to energy saving whilst maintaining comfort by the possibility of lowering the temperature set point. This study also proves that further energy savings can be acquired from the occupants' presence, scheduling, and activities. These factors can increase the comfort probability to more than 10%. [Display omitted] • This paper addresses the gap between the physiology and the psychology thermal comfort approach, dominated by AI solutions. • The work shows a wider comfort zone which has been identified to become progressively narrower over the past several decades. • The field studies represent major UK-dwelling cases that weren't addressed in the previous Artificial Intelligence approach. • The occupant presence and scheduling can contribute to more than a 10% increase in comfort which impacts energy saving. • This work highlights the possibility of achieving indoor thermal comfort with less energy for more sustainable dwellings. [ABSTRACT FROM AUTHOR]

Pipe wall-thinning is a phenomenon whereby the thickness of pipes in a nuclear power plant decreases over time owing to extended operational years. This thickness reduction is caused by various long-term thermal aging mechanisms such as Flow-Accelerated Corrosion, Liquid Droplet Impingement Erosion, cavitation, and flashing. Reducing the thickness of the secondary system pipes to the point of rupture can lead to severe human casualties and significant economic losses. Consequently, domestic power plant operators regularly manage the power plant pipes. Ongoing research focuses on the identification and management of pipe wall-thinning. However, previous studies have encountered problems in accurately judging the decrease in pipe wall-thinning in the presence of noise. To overcome this, Convolutional Neural Network (CNN) models based on image feature analysis have been used. This approach allows for the differentiation of pipe wall-thinning feature from small-sized noise in a single image. However, there were difficulties in making accurate judgments for large-sized noise that resembled the feature of pipe wall-thinning. This paper aims to analyze the limitations of the current pipe wall-thinning evaluation methods and to achieve accurate pipe wall-thinning discrimination through time-series analysis of continuous pipe wall-thinning data. The proposed method employs a Convolutional Recurrent Neural Network (CRNN) model, integrating the Recurrent Neural Network(RNN) model with the CNN model. The image feature of the pipes, extracted using CNN, are utilized as inputs for the RNN. This enables the observation of how the image features of the pipes change over time. This feature differentiates from the time-series feature of noise that occurs suddenly. Through this method, the paper proposes a new approach for effectively identifying the gradual decrease in pipe wall-thinning, enabling precise assessment of pipe wall-thinning progression. [ABSTRACT FROM AUTHOR]

Over the past decades, deep learning techniques have attracted increased attention from various research and industrial domains aligned with the development of Industry Internet-of-Things(IIoT). Specifically, with the advantage of data-driven methods, industrial organizations are seeking novel proactive strategies supported by analytic models to guarantee the quality of their production by observing degradation or predicting failure ahead of the occurrence of the component or asset. Predictive strategies are expected to promise the influence of unnecessary maintenance interruptions and mitigate the consequence of that, hence, extending the remaining useful life of products. This paper conducts a survey of the utilization of deep learning technologies on engineering applications where they provide satisfactory solutions with respect to specific data types or input signals. 106 primary papers are reviewed on deep learning–driven approaches which mainly explore five of the most popular architectures in the application of predictive maintenance. The main content of this paper summarizes the common advantages of each architecture and, accordingly, points out their limitations, as well as describes the application scopes of fully connected deep neural networks, convolutional neural networks, stacked autoencoders, deep belief networks, and deep recurrent neural networks. Based on the technique discussion for each of them, we intend to provide a comprehensive understanding and guidance of the appropriate usage of deep learning architectures to devise an effective predictive maintenance strategy for the scientific and industrial developers whose expertise lies in the prior domain knowledge of multi-source isomerization data. Moreover, the main content demonstrated the summarization of the decisive factor by which the incremental stages of the approaches were determined, fundamentally including the dataset specification, feature extraction, and the integration of deep learning approaches. [ABSTRACT FROM AUTHOR]

With the increasing complexity of the power grid, the precision of the thermal power units' execution of automatic generation control (AGC) commands is gradually increasing the impact on the online regulation of the power grid. The deviation between the actual output of thermal power units and the AGC command of the grid will not only affect the consumption of new energy output, but also endanger the safe operation of the grid. This paper introduces "deep learning" technology to solve the problem. Firstly, an AGC command execution effect identification and confidence evaluation algorithm (ACEEI-CEA) is proposed. The algorithm builds a neural network model to accurately predict the unit output and the confidence evaluation of the prediction results. Next, a high-dimensional input preprocessing strategy based on variational autoencoder (VAE) is proposed to reduce the dimensionality of the model input attributes, improving the convergence and accuracy of the model. Finally, an AGC optimal command fast inversion solution method (AOCFISM) is designed. This method transforms the unit output deviation problem into an objective optimisation problem. And improve the efficiency of solving the optimal AGC command value by constraining the unit command value. The calculation results show that the error of the prediction results of the model proposed in this paper is 5% lower than that of the traditional neural network. The difference between the output value of the optimal AGC command and the expected output value obtained is less than 0.5 MV, which can support AGC online decision-making.© 2017 Elsevier Inc. All rights reserved. • An AGC command execution effect identification and confidence evaluation algorithm (ACEEI-CEA) is proposed. • Results show the ACEEI-CEA performs well in command prediction compared to the state-of-the-art methods. • ACEEI-CEA achieves confidence assessment of predicted values. • The pre-processing strategy based on variational autoencoder shortens the training time of the model. • The AGC optimal command fast inversion solution method (AOCFISM) can obtain the AGC optimal command value. [ABSTRACT FROM AUTHOR]

This paper presents an innovative solution to the key exchange problem in the Industrial Internet of Things (IIoT) implementation. Communication between connected devices must be safe and effective if the IIoT is to continue its fast growth. A crucial part of safeguarding the security and integrity of data shared across the IIoT ecosystem is played by cryptographic key exchange approaches. To provide reliable cryptographic key exchange in the IIoT context, this paper offers a unique approach that makes use of hyperchaotic with complex values, variable parameters, and vector-valued neural synchronization. The method addresses the continued demand for effective cryptographic key exchange across IIoT devices by using drive–response techniques to speed up important applications. Long assessment times in conventional algorithms make it difficult to hide neuronal coordination. As a result, this paper offers a condensed evaluation of ANNs' (Artificial Neural Networks) synchronization by coordinating ANNs for session key exchange using a hyperchaotic setting. The recommended approach has several benefits, including the following: (1) introducing a hyperchaotic system to generate synchronized input vectors for ANN synchronization; (2) The adaptive rules of parameters-based control procedures are constructed mathematically; (3) reciprocal alignment of vector-valued ANNs to form a neural network for establishing session keys throughout the IIoT network; and (4) relevant numerical simulations are carried out to assess the scheme's consistency. The recommended approach performs better than other methods that have been published in the literature, widening up opportunities for more effective and reliable industrial applications. [ABSTRACT FROM AUTHOR]

This paper proposes a real-time state of health (SOH) estimation model based on a deep learning (DL) framework. The proposed model is a combination of two different architectures; a one-dimensional convolutional neural network (1D-CNN) and a bidirectional gated recurrent unit (BiGRU). The hybrid CNN-BiGRU uses the 1D CNN layers to extract pertinent features from input data and then relies on the BiGRU layers for sequence learning in both directions. To account for all SOH indicators, the proposed approach uses the current, voltage, and temperature measurements, which are readily obtainable from the electric vehicle's battery management system (BMS). This prevents the complex and time-consuming feature extraction used in most related papers. Since the hyperparameters have a significant impact on the performance of neural network models, a Bayesian Optimization (BO) technique based on the Gaussian Process (GP) was considered to tune the CNN-BiGRU model hyperparameters. Accordingly, the objective function was able to converge to a low Mean Squared Error (MSE) of 1.2 × 10 − 5 in just 19 iterations. Afterward, to verify the accuracy of the optimized model, a Lithium-ion battery dataset with several discharge profiles provided by the National Aeronautics and Space Administration (NASA) was used. The obtained results demonstrated the accuracy and robustness of the proposed method compared to other commonly used models. The CNN-BiGRU model yielded a Mean Absolute Error (MAE) of 2.080% and a root-mean-square error (RMSE) of 2.516% in the case of the battery set #C, referring to a set with 70 cycles already used at 24 °C. Additionally, the End of life (EOL) indicator error of zero cycles for the same data. [ABSTRACT FROM AUTHOR]

Rail surface defect detection is an essential part of railroad maintenance to prevent safety accidents. Recently, deep learning-based defect detection algorithms have shown impressive detection performance. However, the performance improvement from deep neural networks is based on a strong assumption that the training images have the same data distribution as the test images. This requirement is unrealistic in practical railroad surface defect detection. The appearance of rails varies significantly in different railroad sections due to changes in working conditions and maintenance status. In this paper, we propose a novel uncertainty-inspired unsupervised domain adaptation framework (UIUDA) to improve the generalization of deep learning models on data with distribution differences. Specifically, domain adversarial learning is used to align the feature distributions of training and test data. To mitigate the negative transfer caused by global alignment, we design a locally aligned domain adaptation strategy. In addition, self-training is introduced to refine the decision boundaries of the model in the test data. In this process, we propose an uncertainty-inspired evaluation method to remove noisy pseudo labels. We conducted domain adaptation experiments with RSDD and NEU-RSDD as the source domain and FRSD dataset as the target domain. The results show that UIUDA can effectively improve the generalization ability of the model, and the mIoU metric of defect segmentation is 80.17%. [ABSTRACT FROM AUTHOR]

Electricity demand prediction is crucial to ensure the operational safety and cost-efficient operation of the power system. Electricity demand has predominantly been predicted deterministically, while uncertainty analysis has been usually overlooked. To address this research gap, an integrated Neural Facebook Prophet (NFBP) model and Gaussian Kernel Density Estimation (KDE) model is proposed in this paper, as a way to obtain point and interval predictions of electricity demand, quantifying this way the uncertainty in the predictions. First, historical lagged data, created by utilizing the Partial Auto-correlation Function and Mutual Information Test, is applied to train a prediction model based on NFBP, Deep Learning (DL) as well as Statistical Models. Second, the model Prediction Errors (PE) are derived from the difference between actual and predicted values. A splitting strategy based on the mean and standard deviation of PE is proposed. Finally, electricity demand prediction intervals are obtained by applying Gaussian KDE on split PE. To verify the effectiveness of the proposed model, simulation studies are carried out for three prediction horizons on freely available datasets for the Bulimba sub-station in Southeast Queensland, Australia. Compared with DL models (Long-Short Term Memory Network and Deep Neural Network), the Root Mean Square Error of the NFBP model was reduced by 6.1% and 11.3% for 0.5-hr ahead, 22.7% and 26.3% for 6-hr ahead, and 31.8% and 29.9% for daily prediction. In addition, the Prediction Interval normalized Interval width is smaller in magnitude for the proposed NFBP-KDE model compared to other DL and Statistical models. [Display omitted] • A novel method is proposed for interval prediction of electricity demand. • Kernel density estimation with Gaussian function is applied to obtain interval prediction based on point prediction error. • Parameter of KDE distribution is optimized by minimizing Mean Integrated Square Error. • Deterministic and probabilistic metrics verify the superiority of the proposed prediction model. [ABSTRACT FROM AUTHOR]

The minimization of milling wear from prolonged contact between a tool and a workpiece is crucial in modern manufacturing, where high precision, efficiency, and productivity are required. Therefore, many researchers have developed systems for monitoring milling wear in real-time. However, while the adoption of multiple sensor signals can enhance the training effectiveness of the model, the complex setup process, data extraction, and feature analysis make it challenging to practically apply it for wear monitoring on production lines. Therefore, this study integrates various machining parameters of machine tools and a single type of sensor signal to extract features related to tool wear. These features are utilized as training data for an artificial neural network (ANN) model. The paper proposes a tool wear prediction ANN model based on automatic hyperparameter tuning and transfer learning, incorporating the architecture of Inception, DenseNet, and Layer Normalization. The network structure enhances the training efficiency for target features by expanding and reusing features from a single sensor's data. The combination of automatic hyperparameter tuning with a general optimizer optimizes the ANN architecture to achieve the best tool wear prediction model. Lastly, the proposed model is compared with other models using metrics including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R 2) to evaluate the model performance. In the results, the proposed model shows the best performance among all the comparing models in all three metrics. The MAE is 0.03527, RMSE is 0.05135, and the R 2 is 0.9611. Compared to the common models such as Random Forest, the MAE and RMSE can be reduced by 43.2%, and 43.5% respectively, and a 9.5% R 2 performance increase. • The deep learning model is used for extended feature extraction for the proposed network for transfer learning. • This study proposes a fully connected layers structure which is similar to DenseNet with Layer Normalization at the end of each layer of dense connection. • The model optimization method is used to automatically adjust and optimize hyperparameters. • The experimental results show the proposed model which is capable of predicting tool wear in real-time by feeding the processing parameters has higher prediction accuracy than the other algorithms. [ABSTRACT FROM AUTHOR]

Due to the inevitable light absorption and scattering, underwater images always suffer from severe quality degradation, leading to significant performance decline for various maritime engineering-related applications. In recent years, deep neural networks (DNNs) have been proven to achieve high-quality enhancement of underwater images, which aims to extract and learn abstract features to exhibit superior performance. However, existing solutions often stack multilayer processing units to enrich features, which not only significantly burdens computational load but also encounters difficulties in fully utilizing and interacting between features at different depths. To address the above problems, this paper presents a cost-effective underwater image enhancement network via cascaded feature extraction, termed CFENet. Specifically, we develop a cost-effective cascaded structure for sufficient feature extraction and multiscale establishment of pixel-level long-range dependencies. In addition, we construct a dual-branch structure for effective feature fusion, so that the detailed texture and semantic information in the image can be simultaneously enhanced. Extensive experiments reveal the superiority of our proposed CFENet in both underwater image enhancement effects and computational complexity. Sufficient ablation study is conducted to demonstrate the effectiveness of each component in the network. [ABSTRACT FROM AUTHOR]

Dynamic facial expression recognition (DFER) is of great significance in promoting empathetic machines and metaverse technology. However, dynamic facial expression recognition (DFER) in the wild remains a challenging task, often constrained by complex lighting changes, frequent key-points occlusion, uncertain emotional peaks and severe imbalanced dataset categories. To tackle these problems, this paper presents a depth neural network model based on spatial key-points optimized region feature fusion and temporal self-attention. The method includes three parts: spatial feature extraction module, temporal feature extraction module and region feature fusion module. The intra-frame spatial feature extraction module is composed of the key-points graph convolution network (GCN) and a convolution network (CNN) branch to obtain the global and local feature vectors. The newly proposed region fusion strategy based on face spatial structure is used to obtain the spatial fusion feature of each frame. The inter-frame temporal feature extraction module uses multi-head self-attention model to obtain the temporal information of inter-frames. The experimental results show that our method achieves accuracy of 68.73%, 55.00%, 47.80%, and 47.44% on the DFEW, AFEW, FERV39k, and MAFW datasets. Ablation experiments showed that the GCN module, fusion module, and temporal module improved the accuracy on DFEW by 0.68%, 1.66%, and 3.25%, respectively. The method also achieves competitive results in terms of parameter quantity and inference speed, which demonstrates the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In the process of composite material molding, the role of robots has become increasingly significant. However, the laying process of composite materials is heavily influenced by factors such as temperature and pressure, which affect the quality of the layup. Currently, existing robot path planning algorithms do not account for temperature constraints, presenting challenges in achieving high-quality composite material layups. This paper proposes the RRT*-MTL (Rapidly-Exploring Random Trees* with Minimum Temperature Loss) path planning algorithm. Firstly, a neural network model for predicting temperature loss is developed. Then, temperature loss is integrated as a constraint into the RRT* (Rapidly-Exploring Random Trees*) path planning algorithm to determine the laying path with minimal temperature loss. Experimental results demonstrate that this algorithm effectively preserves the object's heat retention. [ABSTRACT FROM AUTHOR]

In this paper we present a novel methodology, referenced as ORFEO (Ordinal classifier and Regressor Fusion for Estimating an Ordinal categorical target), to enhance the performance in ordinal classification problems for which the latent variable is observable. ORFEO is an artificial neural network model incorporating two outputs, one for ordinal classification, using the cumulative link model, and one for regression, using a linear model. Both outputs are simultaneously optimised considering a loss function that linearly combines both classification and regression losses. The main motivation behind developing the proposed approach is to enhance the performance of a standard ordinal classifier. This improvement is facilitated by considering the regression output, which allows the model to differentiate between patterns within the same category. The ORFEO model is applied to two problems in the field of marine and ocean engineering: short-term prediction of both significant wave height and flux of energy. Both problems are addressed considering four different coastal zones of the United States of America, using 13 datasets formed by buoys measurements and reanalysis data. A comprehensive comparison against 20 methodologies, including regression and nominal/ordinal classification approaches is performed, by using diverse nominal and ordinal performance metrics. Ranks achieved indicate that ORFEO outperforms all the compared methodologies in terms of all the performance measures, demonstrating the efficacy and robustness of the proposal. Finally, a statistical analysis is conducted, concluding that there are statistically significant differences across ordinal and nominal performance metrics in favour of the proposed ORFEO model. [ABSTRACT FROM AUTHOR]

Skin cancer is one of the most serious threats to human health among skin lesions. Computer-aided diagnosis methods can assist patients in identifying and detecting skin lesion types early, thereby enabling corresponding treatments. In this paper, we propose a dual-branch neural network model Conformer with Residual Cosine Similarity Attention and Bidirectional Convolutional fusion strategy, named RCSABC-Conformer. The core of this network structure comprises three parts: a Convolutional Neural Network (CNN) branch with Residual Cosine Similarity Attention (RCSA), a Transformer branch, and a Feature Couple Unit with Bidirectional Convolutional strategy (BC-FCU). The RCSA module calculates the cosine similarity value between the feature map generated by the convolutional operation and the feature map of the residual edge to assess whether their semantic information is similar. The semantic information of similar parts is weighted by exponential normalization to enhance the network's memory of similar features of the same type of skin lesion. The BC-FCU module interactively fuses local features and global representations of skin lesion images with different resolutions in the two branches. Specifically, when the global representations is integrated into local features, we introduce a new bidirectional convolution strategy to extract the feature map from both forward and backward directions, and then select the element with the smaller feature value from the two directions to fuse into local features. In this way, we can minimize the interference of the artifact features extracted by the Transformer branch on the CNN branch. In addition, taking advantage of the Transformer branch's capacity to construct global representations, our model can learn contextual semantic information of normal skin and lesion areas to enhance model robustness. We conducted experiments on three datasets, consisting of clinical and dermoscopic skin lesion images, as well as a hybrid of both. The experimental results show that RCSABC-Conformer outperforms both advanced and classical classification methods in terms of classification accuracy across all three datasets, without requiring an increase in the number of parameters and computational complexity. Compared with the baseline model, the classification accuracy of our proposed method improves by 2.40%, 5.39%, and 4.44% on the three datasets, respectively. To the best of our knowledge, this is the first study to apply an interactive fusion dual-branch network for multi-disease classification on different modalities of skin lesion databases. Code will be available at https://github.com/AlenLi817/RCSABC-Conformer. [ABSTRACT FROM AUTHOR]

The critical heat flux (CHF) is an important parameter determining the heat transfer capability of nuclear reactors. Therefore, prediction of CHF with accuracy and correct understanding is essential for the design and safety analysis of nuclear power plants (NPPs) and industrial boiling systems. This paper focuses on the CHF prediction using the Gaussian process regression (GPR), a machine learning (ML) technique for modeling nonlinear systems. The covariance kernel is a critical component of GPR modeling because it contains assumptions about the function we want to know and defines the dependence and similarity between the data. In this study, the prediction of CHF is investigated during training and prediction in two different approaches: isotropic and anisotropic covariance kernels. Besides, we evaluate the influence of various isotropic and anisotropic kernels on the accuracy of CHF prediction and quantification of uncertainty encoded in the posterior variances in each of the approaches. The use of anisotropic kernels, considering different shape parameters in line with different data dimensions, results in higher accuracy and better generalization. In addition, compared to other ML tools such as support vector regression (SVR) and artificial neural networks (ANNs), the GPR with anisotropic kernels has much better performance. The evaluation criteria involving root mean square error (RMSE) and mean absolute percentage error (MAPE) in anisotropic kernels are significantly improved compared to isotropic kernels. For example, in squared exponential and rational quadratic kernels, RMSE and MAPE values improved by {65.81%, 67.28%} and {61.97%, 60.12%} values, respectively. The results of the GPR method with the anisotropic kernels show the relative superiority of this method compared to SVR and ANN. The lowest RMSE and MAPE values of all three methods were compared, and these values have improved in GPR compared to SVR by 96.01% and 94.25% and compared to ANN by 84.40% and 85.51%, respectively. Finally, some valuable ideas and practical findings will be offered based on numerical simulations and comparisons between isotropic and anisotropic kernel assumptions. Since the approach of using anisotropic kernels has high computational time despite its high flexibility, we will address this issue in future studies and also use this approach for other nuclear engineering problems. [Display omitted] • A flexible GPR structure is presented. • We compared isotropic & anisotropic kernels for critical heat flux prediction. • Anisotropic model outperforms isotropic in critical heat flux prediction. • Anisotropic GPR outperforms SVR & ANN for critical heat flux prediction. • Broader applicability to other CHF predictions in nuclear engineering applications. [ABSTRACT FROM AUTHOR]

Wind speed forecasting plays a crucial role in enhancing the efficiency, reliability, and profitability of renewable energy systems. Accurate wind speed forecasting optimizes energy production and grid integration in renewable energy systems. It assists in maintenance scheduling, decision-making in energy markets, and supports risk management for financial planning. This paper investigates the difficult balance between model complexity and forecasting accuracy in wind speed forecasting using Artificial Neural Networks (ANNs) and Wavelet Neural Networks (WNNs). Employing a time series approach, the study develops models for 1-h-ahead wind speed forecasting, utilizing recent averaged data and exploiting correlations between consecutive wind speeds for improved short-term predictions. Models training involve the Backpropagation algorithm, with careful input variable selection to minimize errors. This study emphasizes that simpler ANN and WNN models can outperform complex ones and explores the effectiveness of wavelet filtering techniques and demonstrates the benefits of using wavelet filtering to enhance forecasting accuracy. The robustness of the developed models is validated through K-fold cross-validation, confirming the efficacy of the proposed models. Evaluation across two datasets demonstrates the superiority of the proposed models over the persistence model, with enhancements of 7–14% in RMSE. Additionally, the efficacy of wavelet transforms in enhancing forecasting accuracy compared to regular neural network predictions is emphasized, underscoring the effectiveness of wavelet filtering offering a strategic approach for optimizing wind power management in electrical grids. [Display omitted] • A single-step-ahead time series wind speed forecasting approach is proposed. • ANN and WNN are employed for wind speed forecasting. • Wavelet filtering is proposed to minimize the noise influence and changes. • Hybrid models for wind speed forecasting using feature selection are proposed. • Daubechies wavelet (db3) is efficient and adopted for wind time series filtering. [ABSTRACT FROM AUTHOR]

Owing to the development of deep learning, indoor personnel detection methods based on deep neural networks have been extensively investigated in recent years. However, more complex and deeper network structures may consume more computational resources, which seriously limits the deployment of large-scale deep neural networks on lightweight devices. In view of this, a lightweight anchor-free one-level feature indoor personnel detection method based on transformer (LAOF-IPDT) is proposed in this paper, which is deployed on two embedded devices and achieves good detection accuracy. In the feature extraction backbone, an enhanced cross-stage partial ghost convolution block with ghost convolution and channel shuffle is designed to extract shallow features. Additionally, to obtain more comprehensive global features in the high-level semantic structure, an embedded vision transformer cross-stage partial block is constructed by embedding a lightweight mobile-friendly vision transformer. For the path-aggregation neck, a lightweight feature pyramid network is proposed, which integrates multi-scale feature maps to obtain richer one-level feature representations. Subsequently, a dilated convolution group block is applied to expand the one-level feature receptive field and detection is accomplished using a one-level feature map. For the detection head, an anchor-free mechanism is applied to reduce the hyper-parameter interference of anchor boxes. Extensive experiments on four datasets indicated that the LAOF-IPDT outperforms other lightweight networks in terms of accuracy, speed, model parameters, and model size. For example, the frames per second of the LAOF-IPDT are 8.39 and 14.67 on CPU devices and Jetson Nano devices, respectively, and the mean average precision is 84.49%. [ABSTRACT FROM AUTHOR]

Monte Carlo Tree Search (MCTS) is a pronounced empirical search algorithm for agent decision-making, especially when enhanced by Deep Learning (DL), in mastering board games that were once thought to be unconquerable. However, it does not appear to be as equally successful in the domain of real-time video games, where the simulation time limit for exploration is a crucial factor, since they are generally designed to be played by human users and hence require a significant amount of resources for simulation. We in this paper propose a surrogate-assisted MCTS approach, specifically targeting commercial real-time video games by approximating the result of gameplay with a deep-learning-based surrogate model. The key contribution of our work is that we designed a modified MCTS for video games that are both commercial and processed in real-time. Since commercial video games include considerably more complex and dynamic gameplays to satisfy their market consumers, as opposed to their non-commercial analogs, our work can be regarded as having challenged the domain unattempted by precedent studies. We validated the performance of our method by conducting a comparative experiment with other algorithms, including the traditional MCTS, under the environment of a commercial real-time video game. [Display omitted] • A Monte Carlo tree search (MCTS) method applicable to commercial real-time video games is proposed. • The method integrates MCTS with a deep neural network structure to function efficiently in a real-time environment. • Performance verification of the proposed surrogate-assisted MCTS is done via two commercial real-time video games with distinct features. • Our surrogate model can be trained with a comparatively little amount of computational resources in real-time video games. [ABSTRACT FROM AUTHOR]

Soil water retention curves (SWRC) are constitutive relationships that are critical for predicting the hydro-mechanical behaviour of unsaturated soil in a wide range of applications in geoenvironmental and geotechnical engineering, hydrology, and agricultural and soil science. Directly measuring SWRC can be costly and time-consuming. Recent research has shown that machine learning, especially artificial neural networks (ANNs), can effectively predict SWRC using easy-to-measure soil properties as inputs. However, significant questions remain about prediction accuracy and what are the most effective input parameters to use, and only shallow neural networks with one hidden layer have been developed in the past for this purpose. In this study, effectiveness of deep neural networks (DNNs) - ANNs with more than one hidden layer - for SWRC prediction is explored for a range of models with different combinations that require easy-to-measure input parameters. Specifically, the study investigates whether switching from a shallow neural network (SNN) to a deep neural network (DNN) can enhance the accuracy of SWRC predictions for these models. Significant differences are found in seven out of thirteen cases when shifting from SNN to DNN. It was also found that models including soil texture, augmented by either dry density or soil porosity as input parameters, perform well. Specific gravity, soil depth, and organic content, on the other hand, do not add to the predictive capability of these models. Findings of this paper can help improve SWRC prediction using machine learning and can lead to more accurate modelling of unsaturated soil behaviour. [ABSTRACT FROM AUTHOR]

The extraction of fault-induced repetitive transients which possess cyclo-stationarity is the key to the fault diagnosis of rotating machinery, which is of considerable significance for ensuring the safe and reliable operation of machinery equipment. Traditional deconvolution methods mainly aim to recover fault-related impulsive features from the time domain and are prone to give poor fault diagnosis results under heavy interference conditions. To solve this problem, a spectrum sparse deep deconvolution method (SSDD) with a deep neural network structure is proposed in this paper. The proposed method uses an envelope spectrum sparse criterion as the cost function to seek an optimal inverse filter through a deep neural network. Firstly, a special band-averaging strategy is designed to initialize the filters in the input layer of the neural network with a window method to provide a direction for deconvolution. Secondly, envelope spectral kurtosis that can depict the sparse feature in the envelope spectrum domain is taken as the cost function to guide the training of the deep network and lock the fault information. Then, the optimal weights are realized by the eigenvalue algorithm, and the weak sparse features are enhanced and extracted layer by layer. Finally, the most significant fault information is obtained through dimension reduction. The simulated and experimental data analysis results verified that the proposed method is superior to traditional deconvolution methods in fault diagnosis performance and robustness to random impulses and strong background noise. [ABSTRACT FROM AUTHOR]

This paper introduces a novel parallel deep neural network for fault diagnosis of drilling pumps. It integrates the Convolutional Block Attention Module with the AlexNet and synchronizes with the Anomaly Transformer model to delve meticulously into both the time and time-frequency domains of signals. The method prioritizes the singular extraction and subsequent amalgamation of features, facilitating a detailed view of diagnostic data and mitigating the risk of interference and overfitting. The integration of the anomaly attention of the Anomaly Transformer with the features of the Convolutional Block Attention Module results in a distinctive dual attention mechanism that is critical to the methodology. This mechanism emphasizes essential features in both the time domain and the time-frequency domain, improving the accuracy of fault diagnosis. Verification with on-site data underscores the preeminence of the approach over existing models, signaling improved reliability and accuracy in diagnosing faults in drilling pumps. This meticulous approach offers promising advances in the study and application of fault diagnosis in energy equipment, demonstrating increased efficiency and refined accuracy. • A novel Parallel Deep Neural Network-Based Fault Diagnosis Framework is proposed. • Time and time-frequency domains of signals are extracted and fused for Fault Diagnosis. • Integrating the Transformer with CBAM to consider the essential features of signals. [ABSTRACT FROM AUTHOR]

Digital image watermarking technology has made a significant contribution to the copyright protection of digital images. In recent years, researchers have focused on designing various watermarking algorithms to enhance resistance against different forms of attacks. However, the evolution of watermarking attack technology has been sluggish, thus impeding possible advancements in digital copyright protection. Existing watermarking attack methods exhibit a notable drawback, causing substantial deterioration in visual quality and undermining the practical utility of attacked images. In this paper, we propose a high imperceptibility watermarking attack network, named HIWANet, based on deep neural networks. To enhance the watermarking attack ability, a feature extraction module (FEM) is aimed to better capture watermark information features, and a watermarking attack module (WAM) is constructed to learn high-level abstract features of the images. In addition, to ensure the imperceptibility of the watermarking attack, an asymmetric loss function is designed to maintain the quality of the attacked watermarked image. In the experiments, we randomly select 2000 color images from the PASCAL VOC2012 database as the dataset, with the training and test sets containing 1000 distinct images. Compared to traditional watermarking attack methods, our HIWANet achieves a significant increase in bit error rate (improved by 242%), indicating a higher attack ability. Meanwhile, it brings more than 26% improvement in the attack imperceptibility. Furthermore, our HIWANet also offers significant advantages compared to deep learning-based watermarking attack methods. [ABSTRACT FROM AUTHOR]

To optimize the operation of his wind farm, the farm manager needs to make precise diagnostic decisions for scheduling efficiently the maintenance actions. To assist him in this task, this paper proposes an approach aimed at designing a hybrid diagnoser model for two main goals. The first one is to detect anomalies at an early stage, and the second one is to provide specific deductions or explanations about the potential cause of the problem. In this way, the developed system can help the maintenance manager understand why a particular decision needs to be made. To do this, an original approach coupling the artificial intelligence technique with a discrete event system is proposed to diagnose, identify, and explain a probable source of a problem. Thus, the proposed method uses a hybrid approach consisting of two model blocks. The first block consists of autoencoder models to extract feature representation to diagnose the health state of the system. The second one is a discrete event-based model to create and visualize rule-based anomaly alerts and triggers to provide plausible explanations to the operator to improve his task. Then, this work introduces a methodology to jointly train these two models and learn all parameters of the hybrid diagnoser. It is shown how the information extracted from the neural network model is used to automatically construct the event-based model to explain the occurrence of an anomaly. The proposed system achieved significant results in explaining and detecting early five types of anomalies in wind turbine systems with accuracy up to 80%. The results demonstrate that the approach can deliver performance gains of up to 20% compared to standard techniques such as Long short-term memory and Random cut forest. • Anomalies detection and explanation in wind turbine systems. • Hybrid learning of an autoencoder and event-based model. • Computation of a set of rules for identifying the root causes of anomalies. • Explainable predictive maintenance for improving decision-making processes. [ABSTRACT FROM AUTHOR]

The issue of Twin Noisy Labels, known as Noisy Annotations and Noisy Correspondences, increases the challenge in the engineering application of Visible–Infrared Person Re-identification (VI-ReID). This paper proposes an novel Modality Blur and Batch Alignment (MBBA) framework to address this issue in the practical application of VI-ReID. The MBBA consists of the Label Confidence Learning (LCL) module, the Modality Blur Learning (MBL) module, and the Batch Alignment Learning (BAL) module. The LCL utilizes the memorization effect of deep neural networks to estimate the confidence level of identity labels and rectify the noisy annotations and the noisy correspondences. The MBL uses the noiseless modality labels and the center loss to blur the boundary among modality data to make the discrimination of the learned latent feature space robust. Based on the designed alignment loss, the BAL employs the self-attention mechanism to align the significant prediction distributions among the cross-modal sample pairs from a batch-size perspective. The mean Average Precision (mAP) is improved by 4.06% and 5.34% compared to state-of-the-art methods on the 20% noisy RegDB dataset. And our MBBA exhibits contemporary state-of-the-art performance on Twin Noisy Labels based VI-ReID. Our MBBA is available at https://github.com/SWU-CS-MediaLab/MBBA. [ABSTRACT FROM AUTHOR]

Segmentation is an important task in a wide range of computer vision applications, including medical image analysis. Recent years have seen an increase in the complexity of medical image segmentation approaches based on sophisticated convolutional neural network architectures. This progress has led to incremental enhancements in performance on widely recognised benchmark datasets. However, most of the existing approaches are computationally demanding, which limits their practical applicability. This paper presents an expand-squeeze dual multiscale residual network (ESDMR-Net), which is a full y convolutional network that is particularly well-suited for resource-constrained computing hardware such as mobile devices. ESDMR-Net focusses on extracting multiscale features, enabling the learning of contextual dependencies among semantically distinct features. The ESDMR-Net architecture allows dual-stream information flow within encoder–decoder pairs. The expansion operation (depthwise separable convolution) makes all of the rich features with multiscale information available to the squeeze operation (bottleneck layer), which then extracts the necessary information for the segmentation task. The Expand-Squeeze (ES) block helps the network pay more attention to under-represented classes, which contributes to improved segmentation accuracy. To enhance the flow of information across multiple resolutions or scales, we integrated dual multiscale residual (DMR) blocks into the skip connection. This integration enables the decoder to access features from various levels of abstraction, ultimately resulting in more comprehensive feature representations. We present experiments on seven datasets from five distinct examples of applications: segmentation of retinal vessels (2 ×), skin lesions (2 ×), digestive tract polyps, lung regions, and cells. Our model demonstrates strong performance, with an F1 score of 0.8287%, 0.8211%, 0.9034%, 0.9451%, 0.9543%, 0.9840%, and 0.8424% on the DRIVE, CHASE, ISIC2017, ISIC2016, CVC-ClinicDB, MC and MoNuSeg datasets, respectively. Remarkably, our model achieves these results despite having significantly fewer trainable parameters, with a reduction of two or even three orders of magnitude. • Multiscale feature extraction to understand contextual dependencies. • The expand-squeeze operation facilitates multiscale information extraction. • Dual-stream information flow within encoder–decoder pairs. • The use of DMR blocks in skip connections improves the information flow. • The deep structure of the network allows it to learn more features. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy (DR) is a prevalent complication of diabetes, affecting a substantial number of individuals worldwide and being a leading cause of blindness. The accurate and automated detection of DR is crucial for effectively managing symptoms such as vision loss and blindness. Recently, there has been significant interest in exploring the applicability of CapsNet for DR grading regarding its success in various vision tasks. However, the performance of traditional CapsNet in DR grading is constrained by the insufficient utilization of capsule features during the training phase. To enhance its performance, this paper proposes a hybrid neural network model called graph neural network (GNN)-fused CapsNet (GF-CapsNet) for DR grading. The model combines various components including ResNet-18 for feature extraction via transfer learning, a PrimaryCaps layer for encoding capsule features, and multi-head prediction that uses GNN-based feature fusion and transformation. Experimental results obtained from two public datasets (Kaggle APTOS 2019 and IDRiD) demonstrate that GF-CapsNet outperforms traditional CapsNet and several other state-of-the-art methods in terms of capturing DR lesions and grading DR. In addition, an investigation into the internal routing process demonstrates that our method mitigates the potential misassignment problem associated with traditional CapsNet. Moreover, the use of the class activation mapping technique for feature map visualization provides an explanation of our model's superior performance in the DR grading task. [ABSTRACT FROM AUTHOR]

Facial image deblurring is an active research area that aims to restore blurry face images to clear ones. However, this task requires special consideration to restore the detailed elements of facial structures, such as eyes, nose, and mouth. Additionally, facial occlusions and varying illuminance conditions in common environments can degrade the deblurring performance. Previous studies have not accounted for these conditions, necessitating the development of deblurring method that considers these factors. In this paper, we propose a novel approach, called Illuminance-robust Multi-stage DeblurNet with Channel Attention (IMDeCA), which leverages semantic mask and landmark information of the face to restore detailed facial structures. Our approach is robust to various illuminance conditions and facial occlusions. The proposed network comprises a multi-stage structure that extracts facial semantic feature maps, reconstructs clear images, and improves illuminance. We also consider facial landmark information in the loss function to ensure well-restored facial structures even in the presence of facial occlusions. Furthermore, we construct a new facial image dataset, named BIO, which includes Blurred images with various types of Illuminance conditions and facial Occlusions. Through extensive experiments on this dataset, we demonstrate the superior performance of our proposed network, outperforming the latest existing methods. [ABSTRACT FROM AUTHOR]

The increasing popularity of mobile applications has catalyzed advancements in object tracking algorithms for mobile platforms. This paper introduces a lightweight object tracking algorithm, designed for real-time and accurate tracking on mobile devices. We propose a feature extraction network with a quantization-friendly architecture and a novel mask-based padding method to enhance tracking precision. By refining MobileNetV3's structure, we mitigate accuracy degradation during quantization and reduce computational demands. The proposed padding technique addresses the outlier generation issue common in traditional padding methods, thereby enhancing positional accuracy. Additionally, an extent adjustment module is incorporated to facilitate prompt error correction during suboptimal tracking. Comprehensive evaluations on OTB50, OTB100, VOT2018, UAV123, and LaSOT demonstrate the superiority of our method over several state-of-the-art trackers in both speed and accuracy. Notably, our feature extraction model is only 3.4 MB, running at 243fps on an RTX 3090 GPU, and can adapt to various sizes of video inputs. • A quantization-friendly structure is designed to reduce the network model size without compromising accuracy. • A mask-based padding method is introduced to better adapt to tracking tasks. • An extent adjustment module is proposed to detect the tracking quality and adjust the size of the search area. • The proposed method achieves excellent tracking accuracy and surpasses similar algorithms in tracking speed. [ABSTRACT FROM AUTHOR]

In weakly coupled few mode fibers (FMFs), used in optical communication systems to increase data capacity, the difference between the effective refractive indices of successive spatial modes is a critical parameter. In this paper, we demonstrate that nanostructured FMFs composed of two different types of subwavelength glass rods can be used to effectively control the propagation properties of individual spatial modes. However, the design and optimization of such fibers proves challenging due to the large number of parameters associated with their free-form refractive index distribution. To address this problem, we use an optimization method based on generative inverse design networks (GIDNs) modified to account for the optical properties of weakly coupled FMFs. Our results show that a nanostructured FMF with a core containing multiple angularly modulated concentric rings can achieve differences between the effective refractive indices of 10 spatial modes exceeding 1.2 × 10−3, surpassing other forms of weakly coupled FMFs. Importantly, the parameters of the structure ensure the feasibility of fabricating the designed fibers. Compared to traditional FMF optimization using a deep neural network with a large, randomly generated data set, our method achieves comparable accuracy in predicting refractive index differences, while achieving better spatial mode separation with a six times smaller data set. By using our proposed approach, which could potentially be applied to the design of other types of optical fibers characterized by a large number of parameters, we can significantly reduce the number of training examples required and the time needed to achieve high-quality results. [ABSTRACT FROM AUTHOR]

Mobile edge computing for autonomous driving needs to manage heterogeneous resources and process large amounts of data or multi-purpose payload. There needs to be deploying, scheduling and migrating tasks on edge nodes to ensure the reliability of tasks or maximize the utilization of resources. However, applying autonomous learning methods on autonomous driving is exceptionally difficult, due to the complexity of multi-dimensional context and the sensitivity to hyperparameters. In this paper, we propose a learning approach to quality-of-service (QoS) prediction of services via multi-dimensional context, and develop a stable approach for service deployment that requires minimal hyperparameter tuning and a modest number of trials to learn multilayer neural network policies. This approach can automatically trades off exploration against exploitation by automatically tuning hyperparameter based on maximum entropy reinforcement learning. We then demonstrate that this approach achieves state-of-the-art performance on Autoware benchmark environments. • This paper is an extension of our icws2017 paper: Xiong, W. , Wu, Z. , Li, B. , & Gu, Q., A Learning Approach to QoS Prediction via Multi-Dimensional Context, 2017 IEEE International Conference on Web Services. We propose a learning approach to quality-of-service (QoS) prediction of services via multi-dimensional context. • We propose a self-adaptive approach for service deployment under Mobile Edge Computing. • We conduct comprehensive experiments on a real-world system, demonstrating the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Abstract Cloud-type recognition is useful in preventing losses caused by adverse weather conditions. This paper presents a methodology and algorithms for automatic recognition of cloud types from color cloud images taken from the ground. We propose a number of eight algorithms for automatic cloud classification of seven cloud types defined by meteorological organization. Recognition is based on a number of features extracted from images which are related to color, texture, and shape. We introduce three new features based on the Fourier transform, namely, the modified k -FFTPX, the half k -FFTPX, and the h × k -FFT. The classification technique is based on artificial neural network (ANN) with tree algorithm to extract features. The proposed classification tree algorithm uses the technique called hierarchical classification which is composed of three levels of tree. Each level of the tree is capable of classifying up to four classes. We show that this method provides the highest accuracy at 98.08% through a series of four experiments. For accuracy assessment, each experiment splits dataset into train and test images using Leave-One-Out Cross-Validation (LOOCV). The result confirms that the hierarchical classification performs better than a single classification. In addition, the tree can be adapted to classify lesser number of cloud types. Our experiment reveals that the accuracy for classifying two classes, cloud and no-cloud, is high as 100%. Highlights • ANN based on tree algorithm is used for classifying 7 cloud types. • Three novel features are proposed based on Fourier transform. • Eight algorithms are presented in the paper. • Four experiments are conducted for testing ANN model on each level of tree. • Our algorithms provide a high accuracy of 98.08%. [ABSTRACT FROM AUTHOR]

This study presents a review of literature on the usage of artificial neural networks (ANNs) architecture contribution method and structural equation modeling (SEM), and proposes a new selection process in the context of algorithm -based SEM-ANNs schemes. This study enriches academic literature by providing a review of all the main aspects of customization in ANNs and contribution methods in combination with SEM. Academic databases are examined for exhibition findings, yielding 253 papers published between 2016 and 2022. The retrieved papers are categorized according to inclusion criteria, and the final set of 73 articles are discussed based on two directions, namely, 'Sector-based' and 'Algorithm-based' as a new representation of taxonomy research. A state-of-the-art bibliographic analysis is presented. This review also identifies modern challenges and open issues in terms of multiple evaluation criteria, importance criteria, and data variations related to the selection of customizations in ANNs and contribution methods combined with SEM in different industrial cases. Several issues fall under multicriteria decision making for handling complexity problems in different ANNs and contribution methods. Thus, this study also presents a research proposal and recommends a solution based on a three-phase methodology for handling the selection and overcoming the identified issues, subsequently completing a strategic guideline solution. [ABSTRACT FROM AUTHOR]

The use of Unmanned Aerial Vehicles (UAVs) has become increasingly popular in recent years, leading to a surge in research on this topic that is widely represented in the literature. They would be examined in accordance with the requirements of the sensors they house, their operating limits, and the output information they give. Since perception, planning, localization, and control are the major tasks of UAVs, an outstanding problem solver "Deep learning" is recently used in these systems. It is ideally suited for UAV applications because of its great capacity for learning representations from the complicated data received in actual situations. This paper provides an extensive review and in-depth analysis of current advancements in UAVs that have applications with DNN, as well as a quick summary of research and development during the previous ten years. In terms of UAVs flight monitoring, remote sensing and vision capability, and energy modeling, this survey provides a roadmap for understanding the sequential development of sophisticated UAVs by reviewing 173 retrieved papers. For this purpose, first of all, the main titles of the studies carried out for UAV applications were gathered under a taxonomy, and then information about the studies carried out in recent years related to these main research fields was given. In the last part of the study, attention was drawn to the still challenging points, such as autonomous fault detection, path planning, and onboard event detection, which are anticipated to be the future trends involving UAV applications. • An up-to-date and comprehensive overview on UAV sensor applications with DNNs. • A taxonomy is given for the researchers to easily examine the studies. • To easily examine the data sets, information about them is summerized. • Information about DNNs architectures end evaluation method is given. • A total of 173 studies, with 117 articles, and 54 conferences proceedings are surveyed. [ABSTRACT FROM AUTHOR]

Safety and passengers' comfort affect directly the reliability and availability of trains. Therefore, appropriate monitoring of the variables influencing those two concepts is mandatory for good operation and maintenance costs reduction. Moreover, operators and rail manufacturers are evolving to a condition-based maintenance strategy (CBM). For this purpose, machine learning is being used to predict early main component's failures, avoiding a loss of reliability or availability during commercial operation. From the passengers' point of view, punctuality and comfortability are two of the most relevant factors when evaluating a railway transport service. Comfort is mainly based on the availability of all services in the train, but it is also related to the accelerations received through the car body to the passengers. More precisely, the lateral car body acceleration (Ayc variable in this paper) is a key variable related to comfort perception and it is therefore important to be monitored and controlled. Rolling, suspension, and damping elements must be kept in perfect conditions during operation to achieve a smooth train movement. In this paper, an optimized model based on Long Short Term Memory networks (LSTM) is designed and trained to predict lateral car body accelerations using only the train accelerometers and with no additional sensoring for this specific experiment. This model will be compared with artificial neural networks (ANN) models to evaluate the relevance of time dependency for this purpose. The optimal time frame to group data is a key characteristic of the neural network training process and is also determined in this paper. The model is trained with data corresponding to the correct dynamic behavior of the train. To test the reliability of this method, a real case of excessive train movement is tested. In these situations, an alert could be sent to the operator or the maintainer for a not-programmed maintenance inspection at the depot (predictive maintenance), avoiding future passenger complaints for the loss of comfort or, in extreme cases, speed limitations due to safety reasons for excessive accelerations. The main contributions of this work are the following: - A LSTM-based neural network, predicting lateral car body accelerations achieving better results than a plain ANN for this dataset: 0.0335 m/s2 MAE and a 12% gain on extreme axles. - An optimization of the time frame to group the measurements in this algorithm that is considering as an input the time variable: 5 s. - A real use case of the model in which monitoring the deviation would have helped to take an early maintenance adjustment on the suspension or damping elements to prevent the loss of comfort. - The p-value used to reduce the uncertainty of the results obtained in the real use case is reaching over 0.95 in the coaches with good dynamic behavior and less than 0.15 in the coaches where a dynamic problem related to comfort loss is detected. [ABSTRACT FROM AUTHOR]

Congenital heart disease affects about 1% of newborns, posing risks like heart failure and mortality. Developing countries often lack resources for diagnosis and treatment. The George B. Moody PhysioNet Challenge 2022 aims to develop systems for detecting murmurs and clinical outcome events using phonocardiogram (PCG) data, offering a cost-effective method for diagnosing cardiac diseases without invasive procedures. We proposed a deep learning model for this task and achieved the 5th rank out of 40 teams in both Track 1 (murmur detection) and Track 2 (clinical outcome prediction). This paper describes our methods and additional experiments. To extract features from the PCG, we employed techniques including Constant Q Transform (CQT) and Mel-scaled spectrogram (Mel-spectrogram) to generate two-dimensional representations in the frequency–time domain. Additionally, we extracted the Peak Interval (PI) feature, which measures the distance between peaks in the PCG data. This feature is useful because the peak intervals should be shorter in PCG recordings with murmurs. We also considered the sequence and mean of PI as additional features. Our proposed system, titled 'Phonocardiogram-based Heart murmur Detection using Spectrogram and PI features (SpectroHeart),' employs the Mel-spectrogram and PI features to detect heart murmurs and assess clinical outcomes. We believe that our deep learning-based system has great potential for automatically detecting heart signals from PCG. • We constructed a deep-learning model for detecting heart murmurs using PCG data. • We experimented with the peak interval sequence and averaged value features. • Achieved the 5th rank in both tracks in the George B. Moody PhysioNet Challenge 2022. [ABSTRACT FROM AUTHOR]

Intelligent reflecting surface (IRS), in which a large number of tunable reflective elements are employed, can enhance the wireless propagation environment in an acceptable manner. Although this issue is operated via intelligently reflecting the electromagnetic waves from the base-station (BS) toward the users, the optimal tuning of the phase shifters at the IRS is arduous. In other words, the passive nature of reflective elements causes some challenges in direct measuring the channels between the IRS, the BS and the users. Even though the conventional methods first estimate the channels and then optimize the system parameters, this paper utilizes an artificial intelligence procedure in which channel estimation has been carried out considering the levenberg-marquardt training algorithm. This is attained via performing some initial processes and trained and estimated each part at different neural network architectures. The estimated output of the system can be expressed via parameters of artificial neural networks which is an important stage for channel estimation. The aforesaid technique is not only making full use of the learning property of neural network, but also by applying this method, there is no need of any matrix computation and the proposed technique is less complex than the traditional pilot-based techniques. Simulation results demonstrate that the offered technique exhibits premiere performance in comparison with conventional common methods. [ABSTRACT FROM AUTHOR]

While Artificial intelligence (AI) has been successfully applied in assessing geotechnical risk, such methods heavily rely on data quality to achieve satisfactory performance, and their results hardly can be interpreted due to their opaque design. With this in mind, this paper aims to address the following research gap: How can we accurately model geotechnical risks using limited data and domain knowledge, and efficiently explain the results of AI model? We develop a physics-guided transfer learning (PGTL) model to enhance the explainability and accuracy of geotechnical risk modeling. With the help of a physical model that simulates the tunnel excavation, a physics-guided dataset with 1000 samples is established and used to train a deep neural network. On these bases, transfer learning is adopted to fuse the features of physics mechanisms and monitoring data, constructing an explainable prediction model of geotechnical risk. To further support risk decision-making, feature relevance techniques are employed to assess the contribution of input parameters to risk. A shield tunnel construction in Wuhan is selected as a case to validate the effectiveness of the proposed method. The PGTL exhibits a more promising accuracy with R 2 of 0.777 in contrast to three popular machine learning approaches, and provides insights into parameters that significantly induce risk, enhancing site managers' understanding of tunnel construction and being conducive to tunnel safety. • A novel explainable artificial intelligence approach for geotechnical risk modeling is proposed. • Incorporate predefined physical guidance to machine learning to enhance data's feature and provide in-model explainability. • Utilize Model-agnostic feature relevance to provide post-hoc explainability without sacrificing predictive ability. • A real tunnel construction in Wuhan, China is taken as a case study for illustration. [ABSTRACT FROM AUTHOR]

Accurate landslide displacement prediction is crucial for prevention and early warning. In this paper, we proposed a novel hybrid multi-step-ahead prediction model (ARIMA-IM) that combines autoregressive integrated moving average (ARIMA) and intelligent models (IMs) for landslide displacement prediction. This model integrates the linear prediction strengths of ARIMA, the signal decomposition capabilities of variational mode decomposition (VMD) and empirical wavelet transform (EWT), and the nonlinear change-capturing ability of IMs. The proposed model cannot only effectively capture abrupt and long-term trends in landslide displacement but also enable high-precision multi-step-ahead prediction. To validate the effectiveness of the proposed model, we predicted the displacement of the Bazimen landslide in the Three Gorges Reservoir by using four IMs, namely long short-term memory (LSTM), bidirectional LSTM (BiLSTM), gate recurrent unit (GRU), and artificial neural network (ANN), in both continuous and jump strategies for multi-step-ahead prediction. Results showed that ARIMA-IM, especially the hybrid model combining ARIMA and deep learning, achieved high accuracy in 1–5-step-ahead prediction. Continuous multi-step-ahead prediction exhibited higher accuracy and provided more prediction information for decision-making. Compared to other models, ARIMA-IM not only achieved multi-step-ahead prediction but also exhibited comparable or better prediction accuracy, which is of high practical significance for landslide disaster early warning. [ABSTRACT FROM AUTHOR]

Recent studies have demonstrated the significant importance of two factors: global context and multi-level semantics, for semantic segmentation models. However, obtaining features that extract these two factors typically results in high computational cost, which poses challenges for unstructured terrain segmentation. In this paper, we propose a Strip and Asymmetric Aggregation Network (SAANet) to collect global context and multi-level semantics while ensuring considerable segmentation accuracy to distinguish secure and navigable areas in wild environments. The SAANet network mainly consists of two essential modules: Global Strip Module (GSM) and Asymmetric Fusion Module (AFM). GSM utilizes four stripe convolutions to capture long-range contextual information while maintaining lower computational complexity. AFM consists of two units: Asymmetric Fusion Unit (AFU) and Residual Aggregation Unit (RAU). AFU based on asymmetric structure leverages attention mechanisms to fuse discriminative semantic clues from different scales, aiming to enhance the recognition of objects with high inter-class similarity and obtain an effective multi-level semantic representation. RAU is used to enhance the significant semantic features from AFU to achieve impressive terrain segmentation results. Extensive experimental results on the Robot Unstructured Ground Driving (RUGD) and Rellis Campus of Texas A&M University (RELLIS) datasets demonstrate that SAANet outperforms other state-of-the-art methods in recognizing outdoor unstructured safe drivable terrain. [ABSTRACT FROM AUTHOR]

Facial expression recognition (FER) is crucial in enhancing human-computer interaction. While current FER methods, leveraging various open-source deep learning models and training techniques, have shown promising accuracy and generalizability, their efficacy often diminishes in real-world scenarios that are not extensively studied. Addressing this gap, we introduce a novel in-the-wild balanced testing facial expression dataset designed for cross-domain validation, called BTFER. We rigorously evaluated widely utilized networks and self-designed architectures, adhering to a standardized protocol. Additionally, we explored different configurations, including input resolutions, class balance management, and pre-trained strategies, to ascertain their impact on performance. Through comprehensive testing across three major FER datasets and our in-depth cross-validation, we have ranked these network architectures and formulated a series of practical guidelines for implementing deep learning-based FER solutions in real-life applications. This paper also delves into the ethical considerations, privacy concerns, and regulatory aspects relevant to the deployment of FER technologies in sectors such as marketing, education, entertainment, and healthcare, aiming to foster responsible and effective use. The BTFER dataset and the implementation code are available in Kaggle and Github , respectively. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles (UAVs), particularly quadrotors, have seen a surge in applications ranging from surveillance to delivery services. However, autonomous control remains a challenge due to the complexity of their dynamics and the need for real-time responsiveness. This paper addresses the significant problem of autonomous quadrotor control by introducing a novel learning method that integrates a deep neural network with advanced control techniques. Novelty of the work is distribution of different learning tasks in different layers of learning and combining the total learning system in a applicable UAV. The purpose is to enhance the quadrotor's ability to perform complex maneuvers autonomously, which is crucial for expanding its application scope. The proposed method begins with the acquisition of direction control using a fractional order proportional–integral–derivative (FOPID) controller. The learning parameters are refined through a fuzzy method, ensuring adaptability and precision. For the optimization of the FOPID parameters, we apply two stochastic algorithms: Genetic Algorithm and Particle Swarm Optimization, comparing their efficacy. Our approach successfully teaches the quadrotor to generate and control a flight path autonomously. The second stage of learning involves proximal policy optimization, enabling the execution of intricate maneuvers. The experimental setup includes a quadrotor on a test stand with three degrees of freedom, where we validate the method through rigorous simulation and physical tests. The results demonstrate that our method not only teaches the quadrotor complex maneuvers but also ensures robust and precise execution, marking a significant advancement in UAV control systems. The implications of this research extend to improving UAV reliability for critical applications, potentially transforming the landscape of autonomous aerial operations. [ABSTRACT FROM AUTHOR]

The inherent nonlinear and memory-dependent input-output characteristics of piezoelectric actuators pose challenges to the precision of piezoelectric positioning systems. In order to solve this problem, this paper firstly transforms the Jiles-Atherton (JA) model into a neural network structure, designs the Jiles-Atherton neural network (JANN), and combines JANN with nonlinear autoregressive exogenous input (NARX) neural network. A hybrid JA-NARX neural network model is proposed for the first time. This model has the advantages of simple structure, high modeling accuracy, and good interpretability. The effectiveness of the proposed JA-NARX neural network model is validated through a series of experiments, specifically assessing its capacity to accurately capture rate-dependent and asymmetric hysteresis characteristics. The results show that although the proposed neural network model has fewer layers and relatively simple structure, it can realize the high-precision modeling of piezoelectric hysteresis dynamics at a lower computational cost. The experimental data shows that, under the excitation of 60 Hz input signal, the model's PV error only accounts for 0.82% of the full scale range, and the modeling performance is far superior to other models. [Display omitted] • A shallow neural network based on Jiles-Atherton differential-equation model is proposed and constructed for the first time. • An innovative JA-NARX hybrid neural network model is proposed by combining Jiles-Atherton with NARX models. • The JA-NARX hybrid neural network model features simple structure, high modelling accuracy, strong interpretability. • The proposed JA-NARX model was rigorously validated through experiments, effectively describing complex dynamic hysteresis. [ABSTRACT FROM AUTHOR]

In this paper, we introduce new, more efficient, methods for training recurrent neural networks (RNNs) for system identification and Model Reference Control (MRC). These methods are based on a new understanding of the error surfaces of RNNs that has been developed in recent years. These error surfaces contain spurious valleys that disrupt the search for global minima. The spurious valleys are caused by instabilities in the networks, which become more pronounced with increased prediction horizons. The new methods described in this paper increase the prediction horizons in a principled way that enables the search algorithms to avoid the spurious valleys. The work also presents a novelty sampling method for collecting new data wisely. A clustering method determines when an RNN is extrapolating, which occurs when the RNN operates outside the region spanned by the training set, where adequate performance cannot be guaranteed. The new method presented in this paper uses a clustering method for extrapolation detection, and then the novel data is added to the original training set. The network performance is improved when additional training is performed with the augmented data set. The new techniques are applied to the model reference control of a magnetic levitation system. The techniques are tested on both simulated and experimental versions of the system. [ABSTRACT FROM AUTHOR]

Existing neural network (NN) models that predict finite element analysis (FEA) of 3D trusses are not generalizable. For example, if a model is designed for a ten-bar truss, it cannot accurately predict the analysis results of a 12-bar truss. Such changes require new sample data and model retraining, reducing the time-saving value of the approach. This paper introduces Generalizable Surrogate Models (GSMs) that use a set of feature descriptors of physical structures to aggregate analysis data from various structures, enabling a more general model that predicts performance for a variety of geometric class, topology and boundary conditions. The paper presents training of generalizable models on parametric dome, wall, and slab structures, and demonstrates the accuracy and generalizability of these GSMs compared to traditional NNs. Results demonstrate first how to combine and use analysis data from various structures to predict the performance of the members of structures of the same class with different topology and boundary conditions. The results further demonstrate how these GSMs more closely predict FEA results than NN models exclusively created for a specific structure. The methodology of this study can be adopted by researchers and engineers to create predictive models for approximation of FEA. [ABSTRACT FROM AUTHOR]

This paper presents the Hierarchical Hopfield Neural Networks (HHNN). HHNN is a novel Hopfield Neural Network (HNN) approach. HHNN is composed of a hierarchy of self-sufficient HNNs, aiming to reduce the neural network structure and mitigate convergence problems. The HNNN composition depends on the applied problem. In this paper, the problem approached is the inter-domain routing for communication networks. Thus, the hierarchy of HNNs mimics the structure of communication networks (domains, nodes, and links). The proof of concept and the comparison between HNNN with the state-of-art HNN occurs using an implementation of them in the Java programming language. Besides, the performance analysis of the HHNN runs on a parallel hardware platform, using VHDL to develop it. The results have demonstrated a reduction of 93 . 75 % and 99 . 98 % in the number of neurons and connections to build the neural network, respectively. Furthermore, the mean time to achieve convergence of HHNN is rough 1 . 52 % of the total time needed by the current state-of-art HNN approach. It is also less susceptible to early convergence problems when used in communications networks with a large number of nodes. Last, but not least, the VHDL implementation shows that convergence time of HHNN is comparable to routing algorithms used in practical applications. [ABSTRACT FROM AUTHOR]

Hyperspectral image classification is an important task in remote sensing image analysis. Traditional machine learning techniques are difficult to deal with hyperspectral images directly, because hyperspectral images have too many redundant spectral channels. In this paper we propose a novel method for hyperspectral image classification, by which spectral and spatial features are jointly exploited from hyperspectral images. Firstly, considering the local similarity in spatial domain, we employ a large spatial window to get image blocks from hyperspectral image Secondly, each spectral channel of the image block is filtered to extract their spatial and spectral features, after that the features are merged by convolutional layers. Finally, the fully-connected layers are used to get the classification result. Comparing with other state-of-the-art techniques, the proposed method pays more attention to the correlation of spatial neighborhood by using a large spatial window in the network. In addition, we combine the proposed network with the traditional support vector machine (SVM) classifier to improve the performance of hyperspectral image classification. Moreover, an adaptive method of the spatial window sizes selection is proposed in this paper. Experimental results conducted on the AVIRIS and ROSIS datasets demonstrate that the proposed method outperforms the state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Stainless steel has many advantages when used in structures, however, the initial cost is high. Hence, it is essential to develop reliable and accurate design methods that can optimize the material. As novel, reliable soft computation methods, machine learning provided more accurate predictions than analytical formulae and solved highly complex problems. The present study aims to develop machine learning models to predict the cross-section resistance of circular hollow section stainless steel stub column. A parametric study is conducted by varying the diameter, thickness, length, and mechanical properties of the column. This database is used to train, validate, and test machine learning models, Artificial Neural Network (ANN), Decision Trees for Regression (DTR), Gene Expression Programming (GEP) and Support Vector Machine Regression (SVMR). Thereafter, results are compared with finite element models and Eurocode 3 (EC3) to assess their accuracy. It was concluded that the EC3 models provided conservative predictions with an average Predicted-to-Actual ratio of 0.698 and Root Mean Square Error (RMSE) of 437.3. The machine learning models presented the highest level of accuracy. However, the SVMR model based on RBF kernel presented a better performance than the ANN, GEP and DTR machine learning models, and RMSE value for SVMR, ANN, GEP and DTR is 22.6, 31.6, 152.84 and 29.07, respectively. The GEP leads to the lowest level of accuracy among the other three machine learning models, yet, it is more accurate than EC3. The machine learning models were implemented in a user-friendly tool, which can be used for design purposes. • The paper studies behaviour of stainless steel CHS stub columns. • ANN, SVR, GEP and DTR machine learning algorithms are developed and validated. • Machine learning-based design approach for predicting the capacity is proposed. • A comparison with EC3 is presented to assess the performance of the models. [ABSTRACT FROM AUTHOR]

Image based view synthesis using deep neural networks provide novel scene views from a set of captured single or multiple images. Multiplane images (MPI) represent scene content as set of RGB α planes within a reference view frustum and render novel views by projecting the content into the target viewpoints. Image based view synthesis with multiple images is very popularly deployed in various areas because it effectively represents geometric uncertainty in ambiguous regions and can convincingly simulate non-Lambertian effects. However, previous image based view synthesis approaches suffer from interpolating and extrapolating information in pixels or ray spaces to generate seamless novel views without occlusion. To effectively improve visual performance for view interpolation and extrapolation, this paper proposes a novel view synthesis with MPI images. From a monocular RGB image, light field images are computationally generated, the proposed depth map guided deep network produces robust MPI using the light field images and their corresponding depth images, and the MPI network embedded with depth attention blocks forces semantic and geometric information to be uniformly distributed and divided among layers. The proposed approach achieves 3.5% and 4.02% improvements in SSIM and PSNR values, comparing to the SOTA approaches. Qualitative analysis on benchmark dataset also verifies the robustness of the proposed approach. [ABSTRACT FROM AUTHOR]

Despite massive development in aerial robotics, precise and autonomous landing in various conditions is still challenging. This process is affected by many factors, such as terrain shape, weather conditions, and the presence of obstacles. This paper describes a deep learning-accelerated image processing pipeline for accurate detection and relative pose estimation of the UAV with respect to the landing pad. Moreover, the system provides increased safety and robustness by implementing human presence detection and error estimation for both landing target detection and pose computation. Human presence and landing pad location are performed by estimating the presence probability via segmentation. This is followed by the landing pad keypoints' location regression algorithm, which, in addition to coordinates, provides the uncertainty of presence for each defined landing pad landmark. To perform the aforementioned tasks, a set of lightweight neural network models was selected and evaluated. The resulting measurements of the system's performance and accuracy are presented for each component individually and for the whole processing pipeline. The measurements are performed using onboard embedded UAV hardware and confirm that the method can provide accurate, low-latency feedback information for safe landing support. • An approach to vision-based detection of landing target for autonomous UAV missions. • Near real-time performance of pipeline is confirmed on modern embedded GPU devices. • Increased safety and robustness by human presence detection and error estimation. • Pipeline's accuracy as measured using the OptiTrack system confirms its capability. [ABSTRACT FROM AUTHOR]

Developing a high-performing traffic incident-duration prediction model is considered a key component for evaluating the impact of these incidents on the roadway network. Various research studies have developed robust incident-duration prediction models. Still, they have faced many issues in providing an accurate prediction result due to the countless data modeling issues, such as complex correlations, highly skewed data distributions, heteroscedasticity, and outliers. This paper investigates the impact of feature optimization (FO) - a relatively new term encompassing two already-known topics: feature engineering (FE) and feature selection (FS) techniques - on the performance of several machine learning models developed for predicting incident durations. The models developed included multivariate linear regression, decision trees, support vector regressors, K-Nearest Neighbors regression, ensembles, and artificial neural networks. Various FO techniques have been used for each model to derive the massive traffic incidents dataset and repeat the prediction process. Our results show that the proposed filtering, wrapper, and embedded FS techniques can successfully reduce the number of features without sacrificing the prediction performance. Using log-normal transformation to deal with continuous data skewness, min-max normalization to deal with data variability, and principal component analysis (PCA) to reform the dataset into a smaller independent feature subset, FE techniques can enhance the accuracy of incident duration estimation over the assessed ML models. The best-performing FE technique was the PCA since performance improvements were observed across all developed ML models. The best-performing FS technique was the Recursive Feature Elimination, outperforming other tested techniques in reducing model complexity while maintaining model accuracy. [ABSTRACT FROM AUTHOR]