Search

• This paper proposes a reasonable sleep stage scoring framework, where we take into account the physiological characteristics of brain waves. • A novel sleep data processing is proposed to automate the sleep scoring task. By wrapping up the sub-networks that are responsible for learning the transient and overall characteristics of EEG during sleep with Transformer, a parallel model for sequence, the performance is improved when compared against the baseline models. • Considering the practicality in a real clinical setting, this paper adopts a mixed-precision strategy to speed up and alleviate the resource cost. Sleep screening is an important tool for both healthcare and neuroscientific research. Automatic sleep scoring is an alternative to the time-consuming gold-standard manual scoring procedure. Recently there have seen promising results on automatic stage scoring by extracting spatio-temporal features via deep neural networks from electroencephalogram (EEG). However, such methods fail to consistently yield good performance due to a missing piece in data representation: the medical criterion of the sleep scoring task on top of EEG features. We argue that capturing stage-specific features that satisfy the criterion of sleep medicine is non-trivial for automatic sleep scoring. This paper considers two criteria: Transient stage marker and Overall profile of EEG features, then we propose a physiologically meaningful framework for sleep stage scoring via mixed deep neural networks. The framework consists of two sub-networks: feature extraction networks, constructed in consideration of the physiological characteristics of sleep, and an attention-based scoring decision network. Moreover, we quantize the framework for potential use under an IoT setting. For proof-of-concept, the performance of the proposed framework is demonstrated by introducing multiple sleep datasets with the largest comprising 42,560 h recorded from 5,793 subjects. From the experiment results, the proposed method achieves a competitive stage scoring performance, especially for Wake, N2, and N3, with higher F1 scores of 0.92, 0.86, and 0.88, respectively. Moreover, the feasibility analysis of framework quantization provides a potential for future implementation in the edge computing field and clinical settings. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has garnered increasing attention as a means to quantitatively analyze the growing and complex medical data to improve individualized patient care. We herein aim to critically examine the current state of ML in predicting surgical outcomes, evaluate the quality of currently available research, and propose areas of improvement for future uses of ML in surgery. A systematic review was conducted in accordance with the Preferred Reporting Items for a Systematic Review and Meta-Analysis (PRISMA) checklist. PubMed, MEDLINE, and Embase databases were reviewed under search syntax "machine learning" and "surgery" for papers published between 2015 and 2020. Of the initial 2677 studies, 45 papers met inclusion and exclusion criteria. Fourteen different subspecialties were represented with neurosurgery being most common. The most frequently used ML algorithms were random forest (n = 19), artificial neural network (n = 17), and logistic regression (n = 17). Common outcomes included postoperative mortality, complications, patient reported quality of life and pain improvement. All studies which compared ML algorithms to conventional studies which used area under the curve (AUC) to measure accuracy found improved outcome prediction with ML models. While still in its early stages, ML models offer surgeons an opportunity to capitalize on the myriad of clinical data available and improve individualized patient care. Limitations included heterogeneous outcome and imperfect quality of some of the papers. We therefore urge future research to agree upon methods of outcome reporting and require basic quality standards. [ABSTRACT FROM AUTHOR]

The current methods of treating organic waste suffer from limited resource usage and low product value. Research and development of value-added products emerges as an unavoidable trend for future growth. Electro-fermentation (EF) is a technique employed to stimulate cell proliferation, expedite microbial metabolism, and enhance the production of value-added products by administering minute voltages or currents in the fermentation system. This method represents a novel research direction lying at the crossroads of electrochemistry and biology. This article documents the current progress of EF for a range of value-added products, including gaseous fuels, organic acids, and other organics. It also presents novel value-added products, such as 1,3-propanediol, 3-hydroxypropionic acid, succinic acid, acrylic acid, and lysine. The latest research trends suggest a focus on EF for cogeneration of value-added products, studying microbial community structure and electroactive bacteria, exploring electron transfer mechanisms in EF systems, developing effective methods for nutrient recovery of nitrogen and phosphorus, optimizing EF conditions, and utilizing biosensors and artificial neural networks in this area. In this paper, an analysis is conducted on the challenges that currently exist regarding the selection of conductive materials, optimization of electrode materials, and development of bioelectrochemical system (BES) coupling processes in EF systems. The aim is to provide a reference for the development of more efficient, advanced, and value-added EF technologies. Overall, this paper aims to provide references and ideas for the development of more efficient and advanced EF technology. [ABSTRACT FROM AUTHOR]

This paper presents a new framework for stochastic updating of a finite element model for a composite plate, considering the influence of temperature on Lamb wave propagation. The framework involves deterministic updating to optimize mechanical properties and stochastic updating to derive probability density functions for key parameters. It utilizes sensitivity analysis and Bayesian inference with Markov-Chain Monte Carlo simulations and the Metropolis–Hastings sampling algorithm. This paper proposes a machine learning surrogate model based on artificial neural networks to improve computational efficiency. This surrogate modeling approach allows parallelized Monte Carlo simulations, reducing updating time significantly without compromising the accuracy of the resulting probability density functions for model parameters. These advancements show a promising way to enhance composite plate modeling and Lamb wave propagation studies, providing a more efficient and accurate approach to verify and validate finite element models with potential applications in engineering simulations. [ABSTRACT FROM AUTHOR]

Using 3D reconstruction techniques within computer vision frameworks can result in more robust and accurate solutions. However, the main challenge lies in the high computation and memory resources required by such methods. To reduce the complexity of these frameworks, a practical solution is to use geometric landmarks instead of the entire image. Therefore, in this paper the problem of 3D shape recovery of a set of standard 2D landmarks, on a single human face image is faced using multi-dimensional scaling (MDS) approach to find a 3D embedding for a set of standard 2D points. Hence, MDS approach is used for the first time in this study to establish an unbiased mapping from 2D landmark space to the corresponding 3D shape space. A deep neural network learns the pairwise 3D dissimilarity among standard 2D landmarks. This scheme leads to find a symmetric dissimilarity matrix, to be fed into the MDS approach to appropriately recovering the 3D shape of corresponding 2D landmarks. In the case of complex input image formations like posedness or perspective projection causing occlusion in the input image, an autoencoder component is used in the proposed framework, as an occlusion removal part, which turns different input views of the human face into a profile view. The results of performance evaluation using variety of synthetic and real-world human face datasets, including NoW dataset, Besel Face Model (BFM), CelebA, CoMA - FLAME, and CASIA-3D, indicates the superiority of the proposed framework, despite its small number of training parameters, with the related state-of-the-art and recent 3D shape recovery of landmark methods from the literature, in terms of unbiasedness, independent projection, efficiency and accuracy. Further, ablation study is performed to find the best training scheme of the deep learning components. All codes and public data of our paper are publicly available for research purposes at https://github.com/s2kamyab/DeepMDS. [ABSTRACT FROM AUTHOR]

In this paper, we are interested in the acceleration of numerical simulations. We focus on a hypersonic planetary reentry problem whose simulation involves coupling fluid dynamics and chemical reactions. Simulating chemical reactions takes most of the computational time but, on the other hand, cannot be avoided to obtain accurate predictions. We face a trade-off between cost-efficiency and accuracy: the numerical scheme has to be sufficiently efficient to be used in an operational context but accurate enough to predict the phenomenon faithfully. To tackle this trade-off, we design a hybrid numerical scheme coupling a traditional fluid dynamic solver with a neural network approximating the chemical reactions. We rely on their power in terms of accuracy and dimension reduction when applied in a big data context and on their efficiency stemming from their matrix-vector structure to achieve important acceleration factors (×10 to ×18.6). This paper aims to explain how we design such cost-effective hybrid numerical schemes in practice. Above all, we describe methodologies to ensure accuracy guarantees, allowing us to go beyond traditional surrogate modeling and to use these schemes as references. • Deep Learning-based hybridization speeds up numerical schemes of atmospheric reentry while maintaining high accuracy. • Initializing a scheme with a hybrid code's prediction reduces the convergence time and keeps the exact same guarantees. • Uncertainty analysis provides statistical guarantees concerning approximation errors when using hybridization code. • Neural network approximation error is statistically lower than many other sources of error inherent to numerical simulations. [ABSTRACT FROM AUTHOR]

The extensive use of Deep Neural Networks (DNNs) encourages people to design domain-specific hardware called Artificial Intelligence (AI) processors. The novel hardware makes optimizations challenging without a proper performance model that reveals working details and performance implications. This paper presents a performance model, Verrocchio, for Huawei DaVinci AI Core, which predicts the execution time of real-world DaVinci kernels. We propose specially-crafted micro-benchmarks to identify contention source, runtime behaviors, and bandwidth sharing, which significantly determine performance. Since DaVinci Core adopts a binary semaphore mechanism for synchronization, Verrocchio views each instruction as a discrete event and manages its execution time based on the programming logic. For evaluation, Verrocchio achieves average error rates of 2.62% and 2.30% in sample kernels for single-core and double-core execution. We demonstrate an optimizing process of matrix multiplications with Verrocchio, achieving speedups of 1.70× for operators and 1.53× for applications and error rates of 5.06% and 5.25%. • Detailed dissections of Huawei DaVinci AI Core, a novel AI processor. • Benchmarking the DaVinci Core bandwidth contention, the key performance factor. • Performance model for accurate execution time prediction of kernel program. • Demonstration of DaVinci kernel optimization and prediction accuracy evaluation. [ABSTRACT FROM AUTHOR]

• We propose novel training algorithms for vocoder-free text-to-speech synthesis using STFT spectra based on generative adversarial networks (GANs). • We demonstrate that using GANs with the original-frequency-resolution amplitude spectra degrades the synthetic speech quality. • We show that the proposed low-frequency-resolution GANs improves the synthetic speech quality. • We also show that using the inverse mel scale for the proposed algorithm further improves the synthetic speech quality. This paper proposes novel training algorithms for vocoder-free text-to-speech (TTS) synthesis based on generative adversarial networks (GANs) that compensate for short-term Fourier transform (STFT) amplitude spectra in low/multi frequency resolution. Vocoder-free TTS using STFT amplitude spectra can avoid degradation of synthetic speech quality caused by the vocoder-based parameterization used in conventional TTS. Our previous work for the vocoder-based TTS proposed a method for incorporating the GAN-based distribution compensation into acoustic model training to improve synthetic speech quality. This paper extends the algorithm to the vocoder-free TTS and propose a GAN-based training algorithm using low-frequency-resolution amplitude spectra to overcome the difficulty in modeling complicated distribution of the high-dimensional spectra. In the proposed algorithm, amplitude spectra are transformed into low-frequency-resolution amplitude spectra by applying an average pooling function along with a frequency axis; then the GAN-based distribution compensation is performed in the low-frequency-resolution domain. Because the low-frequency-resolution amplitude spectra approximately emulate filter banks, the proposed algorithm is expected to improve synthetic speech quality by reducing differences in spectral envelopes of natural and synthetic speech. Furthermore, various frequency scales that are related to human speech perception (e.g., mel and inverse mel frequency scales) can be introduced to the proposed training algorithm by applying an frequency warping function to amplitude spectra. This paper also proposes a GAN-based training algorithm using multi-frequency-resolution amplitude spectra that uses both low- and original-frequency-resolution amplitude spectra to reduce the differences in not only spectral envelopes but also fine structures. Experimental results demonstrate that (1) GANs using low-frequency-resolution amplitude spectra improve speech quality and work robustly against the settings of the frequency resolution and hyperparameters, (2) in comparison among low-, original-, and multi-frequency-resolution amplitude spectra, the use of low-frequency-resolution ones work best improve the synthetic speech quality, and (3) the use of the inverse mel frequency scale for obtaining low-frequency-resolution amplitude spectra further improves synthetic speech quality. [ABSTRACT FROM AUTHOR]

• This paper aims to predict resource allocation accurately and efficiently based on historical requests. Artificial neural networks (ANN) with historically allocated bandwidth from different time series are used to build the prediction model. • This model is constructed by the normal equation method. • The proposed model needs a small dataset for its training phase, while its accuracy is higher than other models. • The discrete dynamical system builds to find an equilibrium point, which is a stable point in the system. • The results show that the system throughput improves by up to 5%, and packet loss improves by up to 15%. Moreover, the prediction accuracy of the system reached 99.48%. Resource allocation is one of the challenges in the time division multiple access (TDM) passive optical network (PON) area. Much research has been done to find an efficient resource allocation method. Using machine learning (ML) and finding proper ML algorithms in predictive dynamic bandwidth allocation remains an open research question. This paper aims to predict resource allocation accurately and efficiently based on historical requests. Artificial neural networks (ANN) with historically allocated bandwidth from different time series are used to build the prediction model. This model is constructed by the normal equation method. The proposed model needs a small dataset for its training phase, while its accuracy is higher than other models. The discrete dynamical system builds to find an equilibrium point, which is a stable point in the system. The proposed model can be implemented as a pre-training, or it can be dynamically refined when the model is running by using some of its output as feedback. The system evaluation verifies that the proposed DBA (EQ-ANN) elevates the quality of service (QoS) metrics, such as system throughput, packet drop probability, and packet means delay. The results show that the system throughput improves by up to 5%, and packet loss improves by up to 15%. Moreover, the prediction accuracy of the system reached 99.48%. [ABSTRACT FROM AUTHOR]

This paper studies the matching problem of cross-modality video data from a discrete distribution alignment view. Central to this discussion is the visible-infrared person re-identification (VI-reID), a crucial feature that bolsters surveillance systems' efficacy in monitoring individuals across diverse lighting conditions. Going beyond traditional image-to-image matching paradigms, a recent study shows that temporal information can bring richer cues to encode the pedestrian representation, improving the representation power of deep neural networks. However, this integration further complicates cross-modality data matching due to the joint processing of spatial and temporal information. This paper formulates the video data as a discrete distribution and aligns the cross-modality video representation by reducing the matching cost between the two distributions. To this end, a natural idea for aligning the videos is to reduce the divergence of distributions. Moreover, the powerful optimal transport (OT) scheme, which generates the optimal matching flows and establishes the relevance of two distributions, is also employed as a way to measure the distance of distributions. Nevertheless, we observe that endowing the OT in the advanced VI-reID feature extractor leads to a non-symmetric measurement. To mitigate this, the paper introduces a new metric, namely symmetric optimal transport (SOT), reformulating OT into a symmetric form. Thorough analyses and empirical studies affirm the superiority of the proposed SOT, which significantly outperforms the current state-of-the-art methods according to standard benchmarking evaluations. [Display omitted] • This paper formally formulates the video sequences into a discrete distribution. • This paper further provides a symmetric optimal transport (SOT) scheme. • Thorough experiments are conducted to evaluate the proposed SOT. [ABSTRACT FROM AUTHOR]

Despite significant amount of work reported in the computer vision literature, segmenting images or videos based on multiple cues such as objectness, texture and motion, is still a challenge. This is particularly true when the number of objects to be segmented is not known or there are objects that are not classified in the training data (unknown objects). A possible remedy to this problem is to utilize graph-based clustering techniques such as Correlation Clustering. It is known that using long range affinities (Lifted multicut), makes correlation clustering more accurate than using only adjacent affinities (Multicut). However, the former is computationally expensive and hard to use. In this paper, we introduce a new framework to perform image/motion segmentation using an affinity learning module and a Message Passing Graph Neural Network (MPGNN). The affinity learning module uses a permutation invariant affinity representation to overcome the multi-object problem. The paper shows, both theoretically and empirically, that the proposed MPGNN aggregates higher order information and thereby converts the Lifted Multicut Problem (LMP) to a Multicut Problem (MP), which is easier and faster to solve. Importantly, the proposed method can be generalized to deal with different clustering problems with the same MPGNN architecture. For instance, our method produces competitive results for single image segmentation (on BSDS dataset) as well as unsupervised video object segmentation (on DAVIS17 dataset), by only changing the feature extraction part. In addition, using an ablation study on the proposed MPGNN architecture, we show that the way we update the parameterized affinities directly contributes to the accuracy of the results. • A new generalized framework to perform image segmentation/ motion segmentation. • Our affinity learning module overcomes the problem of multiple unknown objects. • Message Passing GNN module aggregates long-range information in a faster way. • MPGNN with node and edge updates successfully aggregates long range information. • Our method is efficient and can be generalized to multiple tasks. [ABSTRACT FROM AUTHOR]

In this paper we approach the problem of unique and stable identifiability from a finite number of input-output samples of generic feedforward deep artificial neural networks of prescribed architecture with pyramidal shape up to the penultimate layer and smooth activation functions. More specifically we introduce the so-called entangled weights , which compose weights of successive layers intertwined with suitable diagonal and invertible matrices depending on the activation functions and their shifts. We prove that instances of entangled weights are completely and stably approximated by an efficient and robust algorithm as soon as O (D 2 × m) nonadaptive input-output samples of the network are collected, where D is the input dimension and m is the number of neurons of the network. Moreover, we empirically observe that the approach applies to networks with up to O (D × m L) neurons, where m L is the number of output neurons at layer L. Provided knowledge of layer assignments of entangled weights and of remaining scaling and shift parameters, which may be further heuristically obtained by least squares fitting, the entangled weights identify the network completely and uniquely. To highlight the relevance of the theoretical result of stable recovery of entangled weights, we present numerical experiments, which demonstrate that multilayered networks with generic weights can be robustly identified and therefore uniformly approximated by the presented algorithmic pipeline. [ABSTRACT FROM AUTHOR]

• Unlike common neural network models, we propose a new whole heart segmentation framework based on improved Deep Forest, which is interpretable. It adopts the idea of ensemble learning and combines the methods of bagging and boosting. • We employ hybrid feature fusion, multi-scale fusion and multi-resolution fusion to increase model diversity and segmentation accuracy, and further compare different feature's importance. • Compared to the neural network models, our model can complete training in a shorter time and is more efficient. Automatic whole heart segmentation plays an important role in the treatment and research of cardiovascular diseases. In this paper, we propose an improved Deep Forest framework, named Multi-Resolution Deep Forest Framework (MRDFF), which accomplishes whole heart segmentation in two stages. We extract the heart region by binary classification in the first stage, thus avoiding the class imbalance problem caused by too much background. The results of the first stage are then subdivided in the second stage to obtain accurate cardiac substructures. In addition, we also propose hybrid feature fusion, multi-resolution fusion and multi-scale fusion to further improve the segmentation accuracy. Experiments on the public dataset MM-WHS show that our model can achieve comparable accuracy in about half the training time of neural network models. [ABSTRACT FROM AUTHOR]

In recent years many amazing deep learning models have been developed, but in the process of practical applications, people often find that these deep learning models have high requirements for hardware storage space and computing power. In Artificial Intelligent of Things (AIoT) scenario, the computing power of the edge or terminal side are relatively limited, therefore, most conventional deep learning models are difficult to be deployed into AIoT devices. It is significant to explore the different performance under different scales of deep learning models. In this paper, we mainly propose a method to analyze the impact of deep learning models with various sizes through various experiments. We employ slimmable network as a Neural Archtecture Search (NAS) tool to realize various model size freely, and evaluate them on the indicators of flops, robustness and accuracy. The experimental results show the variation of flops, robustness and accuracy with the various model sizes, which help understand the impact on performance of deep learning models with different scales in AIoT systems. • Interpreting robustness of deep learning models with different model complexity. • Employing slimmable network to compress neural network. • Understanding the relevance between robustness and accuracy of deep learning models. [ABSTRACT FROM AUTHOR]

Convex ℓ 1 regularization using an infinite dictionary of neurons has been suggested for constructing neural networks with desired approximation guarantees, but can be affected by an arbitrary amount of over-parametrization. This can lead to a loss of sparsity and result in networks with too many active neurons for the given data, in particular if the number of data samples is large. As a remedy, in this paper, a nonconvex regularization method is investigated in the context of shallow ReLU networks: We prove that in contrast to the convex approach, any resulting (locally optimal) network is finite even in the presence of infinite data (i.e., if the data distribution is known and the limiting case of infinite samples is considered). Moreover, we show that approximation guarantees and existing bounds on the network size for finite data are maintained. [ABSTRACT FROM AUTHOR]

• A novel physics-informed convolutional neural network f-PICNN for PDEs without any labelled data. • Nonlinear convolutional units (NCUs). • Memory mechanism (numerical results show it can considerably speed up the convergence). • Finite discretization schemes (e.g., Euler, Crank Nicolson and fourth-order Runge-Kutta). • Auto-regressive framework. The physics and interdisciplinary problems in science and engineering are mainly described as partial differential equations (PDEs). Recently, a novel method using physics-informed neural networks (PINNs) to solve PDEs by employing deep neural networks with physical constraints as data-driven models has been pioneered for surrogate modelling and inverse problems. However, the original PINNs based on fully connected neural networks pose intrinsic limitations and poor performance for the PDEs with nonlinearity, drastic gradients, multiscale characteristics or high dimensionality in which the complex features are hard to capture. This leads to difficulties in convergence to correct solutions and high computational costs. To address the above problems, in this paper, a novel physics-informed convolutional neural network framework based on finite discretization schemes with a stack of a series of nonlinear convolutional units (NCUs) for solving PDEs in the space-time domain without any labelled data (f-PICNN) is proposed, in which the memory mechanism can considerably speed up the convergence. Specifically, the initial conditions (ICs) are hard-encoded into the network as the first time-step solution and used to extrapolate the next time-step solution. The Dirichlet boundary conditions (BCs) are constrained by soft BC enforcement while the Neumann BCs are hard enforced. Furthermore, the loss function is designed as a set of discretized PDE residuals and optimized to conform to physics laws. Finally, the proposed auto-regressive model has been proven to be effective in a wide range of 1D and 2D nonlinear PDEs in both space and time under different finite discretization schemes (e.g., Euler, Crank Nicolson and fourth-order Runge-Kutta). The numerical results demonstrate that the proposed framework not only shows the ability to learn the PDEs efficiently but also provides an opportunity for greater conceptual simplicity, and potential for extrapolation from learning the PDEs using a limited dataset. [ABSTRACT FROM AUTHOR]

We propose, in this paper, a Variable Spiking Wavelet Neural Operator (VS-WNO), which aims to bridge the gap between theoretical and practical implementation of Artificial Intelligence (AI) algorithms for mechanics applications. With recent developments like the introduction of neural operators, AI's potential for being used in mechanics applications has increased significantly. However, AI's immense energy and resource requirements are a hurdle in its practical field use case. The proposed VS-WNO is based on the principles of spiking neural networks, which have shown promise in reducing the energy requirements of the neural networks. This makes possible the use of such algorithms in edge computing. The proposed VS-WNO utilizes variable spiking neurons, which promote sparse communication, thus conserving energy, and its use is further supported by its ability to tackle regression tasks, often faced in the field of mechanics. Various examples dealing with partial differential equations, like Burger's equation, Allen Cahn's equation, and Darcy's equation, have been shown. Comparisons have been shown against wavelet neural operator utilizing leaky integrate and fire neurons (direct and encoded inputs) and vanilla wavelet neural operator utilizing artificial neurons. The results produced illustrate the ability of the proposed VS-WNO to converge to ground truth while promoting sparse communication. • Neuroscience inspired operator learning is proposed for scientific computing. • Proposed VS-WNO promotes sparse communications and is energy efficient. • We introduce a tailored spiking loss function to limit spiking activity. • Numerical examples solved illustrate accuracy and energy efficiency of VS-WNO. [ABSTRACT FROM AUTHOR]

Image super resolution involves enhancing the spatial resolution of low-quality images and improving their visual quality. As in many real-life situations, the image degradation process is unknown, performing the task of image super resolution in a blind manner is of paramount importance. Deep neural networks provide high performances for the task of blind image super resolution, in view of their end-to-end learning capability between the low-resolution images and their ground truth versions. Generally speaking, deep blind image super resolution networks initially estimate the parameters of the image degradation process, such as blurring kernel, and then use them for super-resolving the low-resolution images. In this paper, we develop a novel deep learning-based scheme for the task of blind image super resolution, in which the idea of leveraging the hybrid representations is utilized. Specifically, we employ the deterministic and stochastic representations of the blurring kernel parameters to train a deep blind super resolution network in an effective manner. The results of extensive experiments prove the effectiveness of various ideas used in the development of the proposed deep blind image super resolution network. • A transformer-based neural network for blind image super resolution is proposed. • Hybrid representations of blurring kernel are utilized in the network architecture. • Network parameters interpolation technique is employed for reducing the complexity. • Superior blind image super resolution performance is achieved. [ABSTRACT FROM AUTHOR]

On the international level, it is common to act on reducing emissions from energy systems. However, in addition to industrial emissions, low-stack emissions also make a significant contribution. A good step in reducing its environmental impact, is to move to biofuels, including biomass. This paper examines the impact of placing a catalytic system in a retort boiler to minimize emissions of greenhouse gases, dust and other pollutants when burning pellets. The effect of platinum, and oxides of selected metals placed on the deflector as a solid catalyst was studied. Based on the experimental data, a branched artificial neural network was constructed and trained. The routing of three parallel topologies made it possible to achieve high accuracy while keeping the input data relatively simple. The system showed an average error of 3.54% against arbitrary test data. On the basis of experimental data as well as predictions returned by the artificial neural network, recommendations were shown for the catalysts used and their amounts. Depending on the biomass from which the pellet was produced, the experiment suggested the use of titanium or copper oxides. In the case of the neural network, it was able to select a better system, based on platinum, improving emission reductions by up to more than 19%, depending on the type of pellet used. [Display omitted] • Unique three-path-ANN system to simulate catalytic biomass combustion. • High accuracy with MRE of 3.54 % versus unbiased test data. • Bioenergy production from straw, hemp and wood pellets. • Complexed system to forecast 11 different parameters. • Case study indicates optimal catalyst – feedstock pairs. [ABSTRACT FROM AUTHOR]

Convolutional neural networks are data-driven Image Quality Assessment (IQA) models based on the convolutional operation in a rectangular window. Deformable convolutions with learnable receptive fields can efficiently extract structural features of irregular objects in an image. By superimposing deformable convolutions, a plug-and-play module is designed to obtain information for irregular geometric shapes. To selectively fuse shallow and deep features, we propose a weighted non-local attention (WNLA) module with the input and output of self-attention in a weighted manner. This paper proposes a dual branch residual full-reference IQA network that combines weighted non-local attention and stacked deformable convolution. The proposed network was trained on PIPAL dataset and tested on LIVE and TID2013. The cross-dataset evaluation shows that the network has a competitive generalization ability. Ablation experiments indicate that the proposed modules can effectively improve the performance of the network. Comparative experiments reveal that our network is superior to existing excellent networks. The codes for training, test and visualization are available at: https://github.com/Pengchang-haha/SDCN.git. • BRB was designed to compensate for the information loss from deep convolutions. • WNLA is proposed to capture the interaction information between patches. • Deformable convolutions were stacked to extract more comprehensive features. [ABSTRACT FROM AUTHOR]

Accurately predicting RNA-protein binding sites is essential to gain a deeper comprehension of the protein-RNA interactions and their regulatory mechanisms, which are fundamental in gene expression and regulation. However, conventional biological approaches to detect these sites are often costly and time-consuming. In contrast, computational methods for predicting RNA protein binding sites are both cost-effective and expeditious. This review synthesizes already existing computational methods, summarizing commonly used databases for predicting RNA protein binding sites. In addition, applications and innovations of computational methods using traditional machine learning and deep learning for RNA protein binding site prediction during 2018–2023 are presented. These methods cover a wide range of aspects such as effective database utilization, feature selection and encoding, innovative classification algorithms, and evaluation strategies. Exploring the limitations of existing computational methods, this paper delves into the potential directions for future development. DeepRKE, RDense, and DeepDW all employ convolutional neural networks and long and short-term memory networks to construct prediction models, yet their algorithm design and feature encoding differ, resulting in diverse prediction performances. [Display omitted] • Help researchers better understand RNA-binding proteins and associated mechanisms. • Using computational methods to predict related sites is less costly and faster. • Summarize databases, present the applications and innovations from 2018 to 2022. • Compare differences of models in the design of algorithms and feature extracted. • Discuss the shortcomings and potential directions for future development. [ABSTRACT FROM AUTHOR]

This paper presents a new approach for modelling and predicting the evolution of moisture content (MC) in couscous grains during drying using artificial neural networks. Couscous grains with an initial moisture content of 42 ± 1% were dried in an oven using the gravimetric method. Experimental data on moisture content versus drying time were obtained for nine groups of measurements at relative humidities of 7, 32 and 75% at drying temperatures of 30, 40 and 50°C. Different parameters of the ANN model, such as three learning algorithms and four activation functions, as well as the number of neurons in the hidden layer, were tested and evaluated using the correlation coefficient (R2) and the mean squared error (MSE) to find the optimal structure without overfitting. The best performing ANN structure consisted of the log sigmoid function in the hidden layer, with 24 neurons and the linear function in the output layer, trained by the Levenberg-Marquardt backpropagation algorithm. A significant agreement of predictions with experimental data was observed, with an R2 value of 0.99999 and an MSE value of 8.2173 ×10−4. • New ANN model for predicting the moisture content of couscous grains. • Various ANN structures and parameters assessed with MSE and R2. • This ANN excels in predicting couscous moisture with temperature and humidity. [ABSTRACT FROM AUTHOR]

The latest advances in artificial intelligence deep learning models are unprecedented. A wide spectrum of application areas is now thriving thanks to available massive training datasets and gigantic complex neural network models. Those two characteristics demand outstanding computing power that only advanced computing platforms can provide. Therefore, distributed deep learning has become a necessity in capitalizing on the potential of cutting-edge artificial intelligence. Two basic schemes have emerged in distributed learning. First, the data-parallel approach, which aims at dividing the training dataset into multiple computing nodes. Second, the model-parallel approach, which splits layers of a model into several computing nodes. Each scheme has its upsides and downsides, particularly when running on large machines that are susceptible to soft errors. Those errors occur as a consequence of several factors involved in the manufacturing process of current electronic components of supercomputers. On many occasions, those errors are expressed as bit flips that do not cause the whole system to crash, but generate wrong numerical results in computations. To study the effect of soft error on different approaches for distributed learning, we leverage checkpoint alteration, a technique that injects bit flips on checkpoint files. It allows researchers to understand the effect of soft errors on applications that produce checkpoint files in HDF5 format. This paper uses the popular deep learning PyTorch tool on two distributed-learning platforms: one for data-parallel training and one for model-parallel training. We use well-known deep learning models with popular training datasets to provide a picture of how soft errors challenge the training phase of a deep learning model. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) have gained widespread popularity in different domain applications due to their dominant performance. Despite the prevalence of massively parallel multi-core processor architectures, adopting large DNN models in embedded systems remains challenging, as most embedded applications are designed with single-core processors in mind. This limits DNN adoption in embedded systems due to inefficient leveraging of model parallelization and workload partitioning. Prior solutions attempt to address these challenges using data and model parallelism. However, they lack in finding optimal DNN model partitions and distributing them efficiently to achieve improved performance. This paper proposes a DNN model parallelism framework to accelerate model training by finding the optimal number of model partitions and resource provisions. The proposed framework combines data and model parallelism techniques to optimize the parallel processing of DNNs for embedded applications. In addition, it implements the pipeline execution of the partitioned models and integrates a task controller to manage the computing resources. The experimental results for image object detection demonstrate the applicability of our proposed framework in estimating the latest execution time and reducing overall model training time by almost 44.87% compared to the baseline AlexNet convolutional neural network (CNN) model. • Parallel Computing. • Machine Learning. • DNN Model Partitioning. • Embedded Systems. • Embedded Software. [ABSTRACT FROM AUTHOR]

• Lightweight end-to-end neural network model --- Tiny-COUNET with state-of-the-art cough detection performance. • A technique enables continuous cough detection by leveraging the always-on active noise cancellation microphones in commodity hearables. • The first work validating the possibility of cough detection based on consumer hearables. Cough event detection is the foundation of any measurement associated with cough, one of the primary symptoms of pulmonary illnesses. This paper proposes HearCough, which enables continuous cough event detection on edge computing hearables, by leveraging always-on active noise cancellation (ANC) microphones in commodity hearables. Specifically, we proposed a lightweight end-to-end neural network model — Tiny-COUNET and its transfer learning based traning method. When evaluated on our acted cough event dataset, Tiny-COUNET achieved equivalent detection performance but required significantly less computational resources and storage space than cutting-edge cough event detection methods. Then we implemented HearCough by quantifying and deploying the pre-trained Tiny-COUNET to a popular micro-controller in consumer hearables. Lastly, we evaluated that HearCough is effective and reliable for continuous cough event detection through a field study with 8 patients. HearCough achieved 2 Hz cough event detection with an accuracy of 90.0% and an F1-score of 89.5% by consuming an additional 5.2 mW power. We envision HearCough as a low-cost add-on for future hearables to enable continuous cough detection and pulmonary health monitoring. [ABSTRACT FROM AUTHOR]

This paper provides a worked example of using Dynamic Causal Modelling (DCM) and Parametric Empirical Bayes (PEB) to characterise inter-subject variability in neural circuitry (effective connectivity). It steps through an analysis in detail and provides a tutorial style explanation of the underlying theory and assumptions (i.e, priors). The analysis procedure involves specifying a hierarchical model with two or more levels. At the first level, state space models (DCMs) are used to infer the effective connectivity that best explains a subject's neuroimaging timeseries (e.g. fMRI, MEG, EEG). Subject-specific connectivity parameters are then taken to the group level, where they are modelled using a General Linear Model (GLM) that partitions between-subject variability into designed effects and additive random effects. The ensuing (Bayesian) hierarchical model conveys both the estimated connection strengths and their uncertainty (i.e., posterior covariance) from the subject to the group level; enabling hypotheses to be tested about the commonalities and differences across subjects. This approach can also finesse parameter estimation at the subject level, by using the group-level parameters as empirical priors. The preliminary first level (subject specific) DCM for fMRI analysis is covered in a companion paper. Here, we detail group-level analysis procedures that are suitable for use with data from any neuroimaging modality. This paper is accompanied by an example dataset, together with step-by-step instructions demonstrating how to reproduce the analyses. • This guide walks through a group effective connectivity study using DCM and PEB. • It explains recently developed tools for hierarchical Bayesian modelling. • The appendices clarify the technical detail of the PEB framework and its priors. • An accompanying dataset is provided with step-by-step analysis instructions. [ABSTRACT FROM AUTHOR]

• Deep learning accelerates vapor liquid equilibrium prediction with verified accuracy. • Comprehensive comparison between deep learning and various flash calculation methods. • Optimized deep learning model for estimating VLE properties. • Understandable introduction of deep learning and flash calculation for general audience. • Remarks on further applications of deep learning in classical engineering problems. In the past two decades, researchers have made remarkable progress in accelerating flash calculation, which is very useful in a variety of engineering processes. In this paper, general phase splitting problem statements and flash calculation procedures using the Successive Substitution Method are reviewed, while the main shortages are pointed out. Two acceleration methods, Newton's method and the Sparse Grids Method are presented afterwards as a comparison with the deep learning model proposed in this paper. A detailed introduction from artificial neural networks to deep learning methods is provided here with the authors' own remarks. Factors in the deep learning model are investigated to show their effect on the final result. A selected model based on that has been used in a flash calculation predictor with comparison with other methods mentioned above. It is shown that results from the optimized deep learning model meet the experimental data well with the shortest CPU time. More comparison with experimental data has been conducted to show the robustness of our model. [ABSTRACT FROM AUTHOR]

In this paper we address the problem of constructing a feature extractor which combines Mallat's scattering transform framework with time-frequency (Gabor) representations. To do this, we introduce a class of frames, called uniform covering frames , which includes a variety of semi-discrete Gabor systems. Incorporating a uniform covering frame with a neural network structure yields the Fourier scattering transform S F and the truncated Fourier scattering transform. We prove that S F propagates energy along frequency decreasing paths and its energy decays exponentially as a function of the depth. These quantitative estimates are fundamental in showing that S F satisfies the typical scattering transform properties, and in controlling the information loss due to width and depth truncation. We introduce the fast Fourier scattering transform algorithm, and illustrate the algorithm's performance. The time-frequency covering techniques developed in this paper are flexible and give insight into the analysis of scattering transforms. [ABSTRACT FROM AUTHOR]

An increasing amount of malicious code causes harm on the internet by threatening user privacy as one of the primary sources of network security vulnerabilities. The detection of malicious code is becoming increasingly crucial, and current methods of detection require much improvement. This paper proposes a method to advance the detection of malicious code using convolutional neural networks (CNNs) and intelligence algorithm. The CNNs are used to identify and classify grayscale images converted from executable files of malicious code. Non-dominated Sorting Genetic Algorithm II (NSGA-II) is then employed to deal with the data imbalance of malware families. A series of experiments are designed for malware image data from Vision Research Lab. The experimental results demonstrate that the proposed method is effective, maintaining higher accuracy and less loss. • A technique for converting a malware binary to an image was introduced. • In this paper, a method based on CNN is used to identify and classify the malicious codes. • An effective data equilibrium approach based on the NSGA-II was designed. • The proposed method was demonstrated through the extensive experiments. [ABSTRACT FROM AUTHOR]

As labour requirements in horticultural become more challenging, automated solutions are becoming an effective approach to maintain productivity and quality. This paper presents the design and performance evaluation of a novel multi-arm kiwifruit harvesting robot designed to operate autonomously in pergola style orchards. The harvester consists of four robotic arms that have been designed specifically for kiwifruit harvesting, each with a novel end-effector developed to enable safe harvesting of the kiwifruit. The vision system leverages recent advances in deep neural networks and stereo matching for reliably detecting and locating kiwifruit in real-world lighting conditions. Furthermore, a novel dynamic fruit scheduling system is presented that has been developed to coordinate the four arms throughout the harvesting process. The performance of the harvester has been measured through a comprehensive and realistic field-trial in a commercial orchard environment. The results show that the presented harvester is capable of successfully harvesting 51.0% of the total number of kiwifruit within the orchard with an average cycle time of 5.5s/fruit. • This paper has reported the design and performance of a novel robotic kiwifruit harvesting system. • Detection rate of 89.6% of reachable kiwifruit with a deep network based vision system. • Measurement of its in-orchard performance shows that it is capable of picking 51.0% of kiwifruit in the three test orchards. • It is estimated that with further development the unit may be capable of harvesting 70% of the kiwifruit [ABSTRACT FROM AUTHOR]

Highlights • DSTC6: Dialog Challenge to improve performance of end-to-end dialog systems using Neural Network models and dialog breakdown detection. • Track 1, End-to-End Goal Oriented Dialog Learning: selection of the best system response. Hybrid Code Network and Memory Network were the best models. • Track 2, End-to-End Conversation Modeling: system response generation. 78.5% of the automatically generated sentences were rated as acceptable responses by humans. • Track 3, Dialogue Breakdown Detection. The submitted systems performed as well as humans in detecting dialog breakdown, for both English and Japanese data-sets. This paper describes the experimental setups and the evaluation results of the sixth Dialog System Technology Challenges (DSTC6) aiming to develop end-to-end dialogue systems. Neural network models have become a recent focus of investigation in dialogue technologies. Previous models required training data to be manually annotated with word meanings and dialogue states, but end-to-end neural network dialogue systems learn to directly output natural-language system responses without needing training data to be manually annotated. Thus, this approach allows us to scale up the size of training data and cover more dialog domains. In addition, dialogue systems require a meta-function to avoid deploying inappropriate responses generated by themselves. To challenge such issues, the DSTC6 consists of three tracks, (1). End-to-End Goal Oriented dialogue Learning to select system responses, (2). End-to-End Conversation Modeling to generate system responses using Natural Language Generation (NLG) and (3). Dialogue Breakdown Detection. Since each domain has different issues to be addressed to develop dialogue systems, we targeted restaurant retrieval dialogues to fill slot-value in Track 1, customer services on Twitter by combining goal-oriented dialogues and ChitChat in Track 2 and human-machine dialogue data for ChitChat in Track 3. DSTC6 had 141 people declaring their interests and 23 teams submitted their final results. 18 scientific papers were presented in the wrap-up workshop. We find the blending end-to-end trainable models associated to meaningful prior knowledge performs the best for the restaurant retrieval for Track 1. Indeed, Hybrid Code Network and Memory Network have been the best models for this task. In Track 2, 78.5% of the system responses automatically generated by the best system were rated better than acceptable by humans and this achieves 89% of the number of the human responses rated in the same class. In Track3, the dialogue breakdown detection technologies performed as well as human agreements, in both data-sets of English and Japanese. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we address the problem of estimating a 3D human pose from a single image, which is important but difficult to solve due to reasons, such as self-occlusions, wild appearance changes, and inherent ambiguities of 3D estimation from a 2D cue. These difficulties make the problem ill-posed, which have become requiring increasingly complex estimators to enhance the performance. On the other hand, most existing methods try to handle this problem based on a single complex estimator, which might not be good solutions for 3D human pose estimation. In this paper, to resolve this issue, we propose a multiple-partial-hypothesis-based framework for the problem of estimating 3D human pose from a single image, which can be fine-tuned in an end-to-end fashion. We first select several joint groups from a human joint model using the proposed sampling scheme, and estimate the 3D pose of each joint group separately based on deep neural networks. After that, the estimated poses are aggregated to obtain the final 3D pose using the proposed robust optimization formula. The overall procedure can be fine-tuned in an end-to-end fashion, resulting in better estimation performance. In the experiments, the proposed framework shows the state-of-the-art performances on popular benchmark data sets, namely Human3.6M and HumanEva, which demonstrate the effectiveness of the proposed framework. Highlights • A multiple-partial-hypothesis-based scheme for the single-image-based 3D human pose estimation is proposed. • A novel domain conversion layer which transforms a heatmap to corresponding 2D point coordinates is proposed. • The proposed method can be end-to-end fine-tuned to improve the overall performance. • The proposed framework provides the state-of-the-art performance for popular benchmark data sets. [ABSTRACT FROM AUTHOR]

Highlights • An advanced end to end conversation system for the 6-th edition of Dialog System Technology Challenge (DSTC6). • Applying sequence adversarial training with extension of the objective function to improve both objective and subjective evaluation metrics. • Minimum Bayes risk (MBR) based system combination of multiple neural conversation models. • Example-based response selection using an embedding-based context similarity. • Thorough evaluation of three different neural conversation models, training techniques, and decoding strategies using a help-desk dialog task in DSTC6. Abstract This paper presents adversarial training and decoding methods for neural conversation models that can generate natural responses given dialog contexts. In our prior work, we built several end-to-end conversation systems for the 6th Dialog System Technology Challenges (DSTC6) Twitter help-desk dialog task. These systems included novel extensions of sequence adversarial training, example-based response extraction, and Minimum Bayes-Risk based system combination. In DSTC6, our systems achieved the best performance in most objective measures such as BLEU and METEOR scores and decent performance in a subjective measure based on human rating. In this paper, we provide a complete set of our experiments for DSTC6 and further extend the training and decoding strategies more focusing on improving the subjective measure, where we combine responses of three adversarial models. Experimental results demonstrate that the extended methods improve the human rating score and outperform the best score in DSTC6. [ABSTRACT FROM AUTHOR]

Highlights • Different from previous studies that use a single label as the gold-standard to decide dialogue breakdown, we focused on inherent subjectivity when people regard dialogue breakdowns and variationality due to the design of conversational agents. • We propose methods for dialogue breakdown detection that first cluster annotators based on their label distributions, and then ensemble detectors trained by different sets of annotators. • Two kinds of neural networks are used as detectors. We assume that breakdowns occur locally (word and phrase levels) and/or globally (sentence and context levels). The convolutional neural network (CNN) is applied to detect former type of breakdowns, and long short-term memory (LSTM) detects the latter type of breakdowns. • Our best model achieves 63.6% F-score on breakdown detection, which is 5.6% higher than the baseline using the conditional random fields. • This is an extended version of our workshop paper submitted to the Dialogue Breakdown Detection Challange 3 (DBDC3). We newly conducted additional experiments that confirm effectiveness of clustering annotators based on their annotation distributions. In addition, detailed error analysis revealed characteristics of the CNN and LSTM on the dialogue breakdown detection task. Abstract Dialogue breakdown is a significant problem in conversational agents. Timely breakdown detection helps the agents quickly recover from mistakes, minimizing the impact on user experience. In this paper, we focus on two problems: variations in determining a response that breakdowns a conversation i.e. , subjectivity, and variations in breakdown types due to designs of conversational agents, i.e. , variationality. To address the subjectivity, which decreases the agreement rate among annotators, our methods detect a dialogue breakdown by ensembling detectors trained by different sets of annotators that are grouped using a clustering algorithm. To address the variationality, our methods apply two types of detector architectures to capture global and local breakdowns. The long short-term memory detector considers the global context and the convolutional neural networks detector is sensitive to the local characteristics. The ensemble of all detectors makes a final judgment. The results of the Japanese task in the Dialogue Breakdown Detection Challenge 3 (DBDC3) confirm that our approach significantly outperforms the baseline, which uses the conventional conditional random field. Detailed error analysis reveals that our encoders based on a convolutional neural network and a long short-term memory have different characteristics. It also confirms the effects of annotator clustering. [ABSTRACT FROM AUTHOR]

Highlights • We propose a novel framework to person re-identification. • We combine Gaussian features with deep semantic features to enhance the discrimination of overall features. • Two identification losses and a verification loss are used to our fusion model. • Experimental results on benchmarks achieve the state-of-the-art performance of person re-identification. Abstract Person re-identification is an emerging research field in computer vision. Our paper aims to study how to improve the discrimination of person features. We find that some peculiarities of people have not been better attention in the semantic features of deep learning. However, some features obtained by traditional methods can better express the color, and these features are an important clue for re-identification. Therefore, in this paper, we combine traditional Gaussian features with deep semantic features to enhance the discrimination of overall features. At last, we have achieved good performance on two public datasets (Market1501 and VIPeR) in three main distance method learning (DML). In addition, we applied this model to the task of vehicle re-identification. Experiments show that our method has a great improvement on the VeRi vehicle dataset. We compare the results with the current high level results, which indicates the effectiveness of our model. [ABSTRACT FROM AUTHOR]

Abstract In the past two decades, significant advances have been made on automated electroencephalogram (EEG)-based diagnosis of epilepsy and seizure detection. A number of innovative algorithms have been introduced that can aid in epilepsy diagnosis with a high degree of accuracy. In recent years, the frontiers of computational epilepsy research have moved to seizure prediction, a more challenging problem. While antiepileptic medication can result in complete seizure freedom in many patients with epilepsy, up to one-third of patients living with epilepsy will have medically intractable epilepsy, where medications reduce seizure frequency but do not completely control seizures. If a seizure can be predicted prior to its clinical manifestation, then there is potential for abortive treatment to be given, either self-administered or via an implanted device administering medication or electrical stimulation. This will have a far-reaching impact on the treatment of epilepsy and patient's quality of life. This paper presents a state-of-the-art review of recent efforts and journal articles on seizure prediction. The technologies developed for epilepsy diagnosis and seizure detection are being adapted and extended for seizure prediction. The paper ends with some novel ideas for seizure prediction using the increasingly ubiquitous machine learning technology, particularly deep neural network machine learning. Highlights • In recent years, the frontiers of computational epilepsy research have moved to seizure prediction. • This is a review of research efforts on automated seizure prediction. • Technologies developed for epilepsy diagnosis and seizure detection are adapted and extended for seizure prediction. • Some novel ideas for seizure prediction using deep neural network machine learning are discussed. [ABSTRACT FROM AUTHOR]

• A data engineering pipeline is proposed to sculpt the data for DNNs with domain knowledge. • The spatial information reconstruction approach is proposed to associate mechanical fault with its spatial position. • A comprehensive evaluation on the performance of DNNs trained by different input data has been applied in two experimental cases. As fault diagnosis based on deep neural network (DNN) has been developing from laboratory to industrial application, creating the appropriate data pipeline to train and evaluate the models has become a noticeable challenge. For mechanical fault diagnosis, high-quality data should be consistent with comprehensive fault information. In this paper, a two-stage data quality improvement strategy is proposed to sculpt the data for deep learning models in fault severity estimation. In the first stage, the spatial information reconstruction (SIR) approach is developed to transform the sensing data from time domain into spatial domain to establish the relationship between fault symptom and spatial position. In the second stage, spatial domain signals are organized based on fault characteristics and types of DNN models, which aims to convert data into useful information. The proposed strategy has been verified by experimental cases, and a comprehensive evaluation on the performance of DNNs trained by different input data has been applied. The experimental results proved that with thoughtfully engineered data, baseline DNN models can achieve high accuracy and robustness under different speed conditions. This paper provides a way for engineering practitioners to design industrial DNN applications with their domain knowledge. [ABSTRACT FROM AUTHOR]

This paper focuses on predicting predawn leaf water potential through a self-learning artificial intelligence (SL-AI) algorithm, a novel spectral processing algorithm that is based on the search for covariance modes, providing a direct relationship between spectral information and plant constituents. The SL-AI algorithm was applied in a dataset containing 847 observations obtained with a handheld hyperspectral spectroradiometer (400–1010 nm), structured as: three grapevine cultivars (Touriga Nacional, Touriga Franca and Tinta Barroca), collected in three years (2014, 2015 and 2017), in two test sites in the renowned Douro Wine Region, northeast of Portugal. The Ψ pd SL-AI quantification was tested both in regressive (R2 = 0.97, MAPE = 18.30%) and classification (three classes; overall accuracy = 86.27%) approaches, where the radiation absorption spectrum zones of the chlorophylls, xanthophyll and water were identified along the vegetative growth cycle. The dataset was also tested with Artificial Neural Networks with Principal Component Analysis (ANN-PCA) and Partial Least Square (PLS), which presented worse performance when compared to SL-AI in the regressive (ANN-PCA - R2 = 0.85, MAPE = 43.64%; PLS - R2 = 0.94, MAPE = 28.76%) and classification (ANN-PCA - overall accuracy: 72.37%; PLS - overall accuracy: 73.79%) approaches. The Ψ pd modelled with SL-AI demonstrated, through hyperspectral reflectance, a cause-effect of the grapevine's hydric status with the absorbance of bands related to chlorophyll, xanthophylls and water. This cause-effect interaction could be explored to identify cultivars and cultural practices, hydric, heating and lighting stresses. [Display omitted] • VIS-NIR spectroscopy was used to predict predawn leaf water potential (Ψ pd). • Benchmark of SL-AI against state-of-the-art chemometrics methods in modelling Ψ pd. • Chlorophyll, xanthophyll, and water absorbance bands related to the Ψ pd. • Cause-effect of plant response to abiotic stresses explained through VIS-NIR bands. [ABSTRACT FROM AUTHOR]

In this paper, we describe the development and testing of an active acoustic sensing (AAS) touch interface. The interface detects subtle changes in the vibrational characteristics of an elastic panel when a user applies a small force to the surface at different locations. The system consists of a panel with an affixed force exciter and vibration sensor to monitor the panel vibrations. An automated system was used to manipulate a stylus to apply a light force to the panel at an array of known locations and the vibrational response of the panel was recorded by the affixed sensor. We employed a rectangular grid of 414 points with 10 mm spacing on a 2 mm thick acrylic panel of dimensions 18 cm by 23 cm. Features of the recordings were employed as training data for a deep neural network. The results demonstrate the viability of the AAS interface, and they show the relative performance of the system as a function of the selected features. The demonstration platform achieved a classification accuracy of 100% and a mean distance error of 0.20 mm for regression. The best-performing feature sets were those that contained sufficient spectral resolution to discriminate subtle changes in the center frequencies and amplitudes of the panel's modal resonances in response to small changes in touch location. The AAS touch interface is a practical and inexpensive means to provide accurate touch sensing for large surfaces, such as displays, televisions, and information kiosks. • Prototype vibration-based touch sensor presented. • Employed machine learning techniques informed by analysis of physical system. • Achieved 100 % classification accuracy at 10 mm resolution. [ABSTRACT FROM AUTHOR]

The objective of the relation classification task is to extract relations between entities. Recent studies have found that R-BERT (Wu and He, 2019) based on pre-trained BERT (Devlin et al., 2019) acquires extremely good results in the relation classification task. However, this method does not take into account the semantic differences between different kinds of entities and global semantic information either. In this paper, we set two different fully connected layers to take into account the semantic difference between subject and object entities. Besides, we build a new module named Concat Module to fully fuse the semantic information among the subject entity vector, object entity vector, and the whole sample sentence representation vector. In addition, we apply the average pooling to acquire a better entity representation of each entity and add the activation operation with a new fully connected layer after our Concat Module. Modifying R-BERT, we propose a new model named BERT with Global Semantic Information (GSR-BERT) for relation classification tasks. We use our approach on two datasets: the SemEval-2010 Task 8 dataset and the Chinese character relationship classification dataset. Our approach achieves a significant improvement over the two datasets. It means that our approach enjoys transferability across different datasets. Furthermore, we prove that these policies we used in our approach also enjoy applicability to named entity recognition task. • We take into account the semantic difference between subject and object entities. • We fully fuse the semantic information between entities and the whole sentence. • These policies we used enjoy high applicability to different NLP tasks. [ABSTRACT FROM AUTHOR]

Email data has some characteristics that are different from other social media data, such as a large range of answers, formal language, notable length variations, high degrees of anomalies, and indirect relationships. The main goal in this research is to develop a robust and computationally efficient classifier that can distinguish between spam and regular email content. The benchmark Enron dataset, which is accessible to the public, was used for the tests. The six distinct Enron data sets we acquired were combined to generate the final seven Enron data sets. The dataset undergoes early preprocessing to remove superfluous sentences. The proposed model Bidirectional Long Short-Term Memory (BiLSTM) apply spam labels and to examine email documents for spam. On seven Enron datasets, DNN-BiLSTM performs better than other classifiers in the performance comparison in terms of accuracy. DNN-BiLSTM and convolutional neural networks demonstrated that they can classify spam with 96.39 % and 98.69 % accuracy, respectively, in comparison to other machine learning classifiers. The risks associated with cloud data management and potential security flaws are also covered in the paper. This research presents hybrid encryption as a means of protecting cloud data while preserving privacy by using the hybrid AES-Rabit encryption algorithm which is based on symmetric session key exchange. [ABSTRACT FROM AUTHOR]

Acoustic emission (AE) is widely used for identifying source mechanisms and the deformation stage of steel material. The effectiveness of this non-destructive monitoring technique heavily depends on the quality of the measured AE signals. However, the AE signals from deformation are easily contaminated by the signals from noise in a noisy environment. This paper presents a hybrid model for deformation stage identification, which combines a self-adaptive denoising technique and an Artificial neural network (ANN). In pursuit of model generality, AE signals were collected from tensile coupon tests with various steel materials and loading speeds. First, a decomposition-based denoising method is applied based on the singular spectral analysis (SSA) and variational mode decomposition (VMD), which is defined as SSA-VMD. Its effectiveness is demonstrated by simulated signals and experimental results. Following the use of the denoising technique, an ANN is constructed to identify the deformation stage of steel materials with the input of features extracted from the filtered AE signals. The results indicate that the ANN achieves a high prediction accuracy of 0.93 in the test set and 0.87 in unseen data. By applying this denoising method, the ANN-based approach enables accurate correlation of the collected AE signals to deformation stages. The finding can be used as the basis for the creation of new methodologies for monitoring structural health status of in-service steel structures. [ABSTRACT FROM AUTHOR]

[Display omitted] • A DNN model permanently records the multiple regeneration effects for real-time forecasting of drill bit operating status. • The forecasting accuracy, convergence speed and stability of this model are higher than other time-series forecasting models. • Accurately forecasting the chaotic behavior of the drill bit. • Increasing batch size and output window size decreases forecasting accuracy. • Appropriate input window size can improve forecasting accuracy. The multiple regenerative effects during the drilling process cause the drilling system to exhibit complex motions, even leading to chaotic behavior. Additionally, due to the inability to obtain real-time rock parameters, traditional drilling dynamics models and time-series forecasting models cannot accurately forecast the real-time operational status of the drilling system. In this paper, a deep neural network (DNN) model integrated with rock surface morphology model is proposed to forecast the real-time operational status of the drill bit. This new model can permanently record the changes in rock surface morphology and multiple regenerative effects during the drilling process, enhancing the perception of rock parameters and improving the forecasting accuracy of the drill bit's operational status. Experimental results show that the new model forecasts the drill bit's operational status with higher accuracy compared to other typical time-series forecasting models without integrating surface morphology model. Furthermore, this new model demonstrates superior convergence speed, stability, and generalization ability compared to other models. In addition, the influence of hyperparameters on the forecast accuracy of the new model has also been studied. These results play an important role in forecasting and preventing adverse vibrations such as stick slip, impact, reversal, and rebound of the drill bit. [ABSTRACT FROM AUTHOR]

Authentication of geographical indication (GI) is important for the protection of GI rice due to its unique quality characteristics and high economic value in the market. This paper compared the appearance quality, cooking quality and nutritional quality of four types of GI rice products (Luoding rice, Zhefang Tribute rice, Xijiang rice and Wanbaozhen rice) in China. Luoding rice was found to have the highest nutritional value with respect to protein and amino acid profile. The vitamin content of Xijiang rice was significantly higher than that of other types of GI rice. Principal component analysis showed that Luoding rice had a quite different protein, aspect ratio and amino acid profile from other GI rice samples. Zhefang Tribute rice had quite different cooking quality, fatty acid and sugar profile from Xijiang rice and Wanbaozhen rice. Geographical origin was predicted by using a multilayer perceptron artificial neural network model with quality characteristics as the input variables. It was found that all the samples were accurately predicted in both training and test sets, indicating the potential of using quality characteristics to discriminate GI rice samples from different geographical origins. • Appearance, cooking, nutritional quality of four GIs were investigated and compared. • LD possessed different protein, aspect ratio and amino acid profile from other GI rice. • ZF possessed different cooking quality and fatty acid, sugar profile from XJ and WBZ. • GI rice was predicted using quality characteristics as input variables. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) have gained widespread adoption in various fields; however, their computational cost is often prohibitively high due to the large number of layers and neurons communicating with each other. Furthermore, DNNs can consume a significant amount of energy due to the large volume of data movement and computation they require. To address these challenges, there is a need for new architectures to accelerate DNNs. In this paper, we propose novel neuron grouping and mapping methods for 2D-mesh Network-on-Chip (NoC)-based DNN accelerators considering both fully connected and partially connected DNN models. We present Integer Linear Programming (ILP) and simulated annealing (SA)-based neuron grouping solutions with the objective of minimizing the total volume of data communication among the neuron groups. After determining a suitable graph representation of the DNN, we also apply ILP and SA methods to map the neurons onto a 2D-mesh NoC fabric with the objective of minimizing the total communication cost of the system. We conducted several experiments on various benchmarks and DNN models with different pruning ratios and achieved an average of 40-50% improvement in communication cost. • Novel methodologies proposed for efficient 2D-mesh NoC-based DNN accelerators. • ILP and SA-based neuron grouping and mapping methods to minimize communication cost. • The proposed methods can be used for both fully connected and partially pruned DNNs. • Performance validation on diverse DNN benchmark applications. [ABSTRACT FROM AUTHOR]

Shadow detection is undergoing a rapid and remarkable development along with the wide use of deep neural networks. Benefiting from a large number of training images annotated with strong pixel-level ground-truth masks, current deep shadow detectors have achieved state-of-the-art performance. However, it is expensive and time-consuming to provide the pixel-level ground-truth mask for each training image. Considering that, this paper proposes the first unsupervised deep shadow detection framework, which consists of an initial pseudo label generation (IPG) module, a curriculum learning (CL) module and a self-training (ST) module. The supervision signals used in our learning framework are generated from several existing traditional unsupervised shadow detectors, which usually contain a lot of noisy information. Therefore, each module in our unsupervised framework is dedicated to reduce the adverse influence of noisy information on model training. Specifically, the IPG module combines different traditional unsupervised shadow maps to obtain their complementary shadow information. After obtaining the initial pseudo labels, the CL module and the ST module will be used in conjunction to gradually learn new shadow patterns and update the qualities of pseudo labels simultaneously. Extensive experimental results on various benchmark datasets demonstrate that our deep shadow detector not only outperforms the traditional unsupervised shadow detection methods by a large margin but also achieves comparable results with some recent state-of-the-art fully-supervised deep shadow detection methods. • A novel unsupervised deep shadow detection framework is designed. • We design an initial pseudo label generation (IPG) module by taking advantages of the complementarity of multiple traditional unsupervised shadow detection models. • We jointly utilize curriculum learning (CL) and self-training (ST) to gradually reduce the influence of noisy information within the initial pseudo labels on model training. [ABSTRACT FROM AUTHOR]

Detecting fights from videos and images in public surveillance places is an important task to limit violent criminal behavior. Real-time detection of violent behavior can effectively ensure the personal safety of pedestrians and further maintain public social stability. Therefore, in this paper, we aim to detect real-time violent behavior in videos. We propose a novel neural network model framework based on human pose key points, called Real-Time Pose Net (RTPNet). Utilize the pose extractor (YOLO-Pose) to extract human skeleton features, and classify video level violent behavior based on the 2DCNN model (ACTION-Net). Utilize appearance features and inter frame correlation to accurately detect fighting behavior. We have also proposed a new image dataset called VIMD (Violence Image Dataset), which includes images of fighting behavior collected online and captured independently. After training on the dataset, the network can effectively identify skeletal features from videos and locate fighting movements. The dataset is available on GitHub (https://github.com/ChinaZhangPeng/Violence-Image-Dataset). We also conducted experiments on four datasets, including Hockey-Fight, RWF-2000, Surveillance Camera Fight, and AVD dataset. These experimental results showed that RTPNet outperformed the most advanced methods in the past, achieving an accuracy of 99.4% on the Hockey-Fight dataset, 93.3% on the RWF-2000 dataset, and 93.4% on the Surveillance Camera Fight dataset, 99.3% on the AVD dataset. And with speeds capable of reaching 33fps, state-of-the-art results are achieved with faster speeds. In addition, RTPNet can also have good detection performance in violent behavior in complex backgrounds. • The YOLO-Pose extracts human skeletal features with the pose of key points as input, and combines RGB information and human skeletal points information to recognize and classify combat behavior. • The Weight Selection Module (WSM) uses parameters generated from skeletal data as weight coefficients to allocate skeletal information and RGB information. • The Weight Distribution Module (WDM) further corrects the error of weight coefficients, while the Keyframe Weight Allocation (KWA) is responsible for connecting skeleton information and RGB features to generate the final detection result. [ABSTRACT FROM AUTHOR]

Highlights • Our method can model various temporal dynamics embedded in videos. • The representations of our method are endowed with discriminative ability. • The dimension of the learned representation is as same as the original one. The aim of this paper is to model long-term evolution of an action video with temporal multi-scale representation. This task is tough due to huge intra-class variations in motion speed. Most of the existing methods consider evolution modeling and multi-scale feature fusion in two separated phases, which generates sub-optimal representation. To address this issue, this paper proposes a novel method to integrate the evolution modeling and multi-scale representation into a unified framework. The core idea is to introduce a temporal multi-scale smoothing vector, which is used to define how the representations at different temporal scales are combined together for frame smoothing. By formulating the smoothing vector learning, evolution modeling and classifier training jointly, our method can learn a discriminative and flexible representation of multi-scale rather than a single scale or a fixed multi-scale smoothing. Experimental results on three datasets demonstrate the effectiveness of our method. [ABSTRACT FROM AUTHOR]

A complete fault diagnosis for the rolling bearing is proposed in this paper. Variable predictive model class discrimination (VPMCD) is a conventional pattern recognition method; however, in practice, when the fault diagnosis method is applied to small samples or in multi-correlative feature space, the stability of the VPM constructed based on the least squares (LS) method is not sufficient. Based on affinity propagation (AP) clustering, RReliefF, and sequential forward search, the ARSFS is proposed to select the significant subset of original feature set and to reduce the dimension and multiple correlations of the feature space. Further, this paper uses two kinds of Gaussian Neural Network, namely the Radial Basis Function Neural Network (RBF) and the Generalized Regression Neural Network (GRNN), instead of the LS method to construct predictive models of VPMCD, called AOR-VPMCD. Compared with the conventional VPMCD and its improvements, based on sufficient experiments, the entire process presented in this paper can effectively identify the fault of the rolling bearing. [ABSTRACT FROM AUTHOR]

Recently, the deep learning schemes have been well investigated for improving the driving safety and efficiency in the transportation systems. In this paper, a probabilistic model named as CPGN (Collision Prediction model based on GA-optimized Neural Network) for decision-making in the rear-end collision avoidance system is proposed, targeting modeling the impact of important influential factors of collisions on the occurring probability of possible accidents in the Internet of Vehicles (IoV). The decision on how to serve the chauffeur is determined by a typical deep learning model, i.e., the BP neural network through evaluating the possible collision risk with V2I (Vehicle-to-Infrastructure) communication, V2V (Vehicle-to-Vehicle) communication and GPS infrastructure supporting. The proper structure of our BP neural network model is deeply learned with training data generated from VISSIM with multiple influential factors considered. In addition, since the selection of the connection coefficient array and thresholds of the neural network has great randomness, a local optimization issue is readily occurring during the modeling procedure. To overcome this problem and consider the ability to efficiently find out a global optimization, this paper chooses the genetic algorithm to optimize the coefficient array and thresholds of proposed neural network. For the purpose of enhancing the convergence speed of the proposed model, we further adjust the studying rate according to the relationship between the actual and predicated values of two adjacent iterations. Simulation results demonstrate that the proposed collision risk evaluation framework could offer rationale estimations to the possible collision risk in car-following scenarios for the next discrete monitoring interval. [ABSTRACT FROM AUTHOR]