Search

An information model is outlined, which represents an intelligent system of metallographic analysis in the form of a set of subsystems, the interaction of which ensures the performance of metallographic analysis functions. The structure of the information storage subsystem for metallographic analysis is presented. The deployment model of an intelligent metallographic analysis system is proposed and described. The paper outlines the approach to the presentation of an expert subsystem for metallographic quality control of metals based on a neural network. The process of finding a close precedent in metallographic analysis with reference to a multilayer neural network is described. An intelligent metallographic analysis system is described, which based on proposed information model. A specialized software of an intelligent metallographic analysis system is presented. The functioning results of the developed system for processing images of steel microstructures to determine the steel quantitative parameters is presented. [ABSTRACT FROM AUTHOR]

Logistic regression is the industry standard in credit risk modeling. Regulatory requirements for model explainability have halted the implementation of more advanced, non-linear machine learning algorithms, even though more accurate predictions would benefit consumers and banks alike. Deep neural networks are certainly some of the most prominent non-linear algorithms. In this paper, we propose a deep neural network model for behavioral credit rating. Behavioral models are used to assess the future performance of a bank's existing portfolio in order to meet the capital requirements introduced by the Basel regulatory framework, which are designed to increase the banks' ability to absorb large financial shocks. The proposed deep neural network was trained on two different datasets: the first one contains information on loans between 2009 and 2013 (during the financial crisis) and the second one from 2014 to 2018 (after the financial crisis); combined, they include more than 1.5 million examples. The proposed network outperformed multiple benchmarks and was evenly matched with the XGBoost model. Long-term credit rating performance is also presented, as well as a detailed analysis of the reprogrammed facilities' impact on model performance. [ABSTRACT FROM AUTHOR]

Target tracking technology that is based on aerial videos is widely used in many fields; however, this technology has challenges, such as image jitter, target blur, high data dimensionality, and large changes in the target scale. In this paper, the research status of aerial video tracking and the characteristics, background complexity and tracking diversity of aerial video targets are summarized. Based on the findings, the key technologies that are related to tracking are elaborated according to the target type, number of targets and applicable scene system. The tracking algorithms are classified according to the type of target, and the target tracking algorithms that are based on deep learning are classified according to the network structure. Commonly used aerial photography datasets are described, and the accuracies of commonly used target tracking methods are evaluated in an aerial photography dataset, namely, UAV123, and a long-video dataset, namely, UAV20L. Potential problems are discussed, and possible future research directions and corresponding development trends in this field are analyzed and summarized. [ABSTRACT FROM AUTHOR]

As a special case of machine learning, incremental learning can acquire useful knowledge from incoming data continuously while it does not need to access the original data. It is expected to have the ability of memorization and it is regarded as one of the ultimate goals of artificial intelligence technology. However, incremental learning remains a long term challenge. Modern deep neural network models achieve outstanding performance on stationary data distributions with batch training. This restriction leads to catastrophic forgetting for incremental learning scenarios since the distribution of incoming data is unknown and has a highly different probability from the old data. Therefore, a model must be both plastic to acquire new knowledge and stable to consolidate existing knowledge. This review aims to draw a systematic review of the state of the art of incremental learning methods. Published reports are selected from Web of Science, IEEEXplore, and DBLP databases up to May 2020. Each paper is reviewed according to the types: architectural strategy, regularization strategy and rehearsal and pseudo-rehearsal strategy. We compare and discuss different methods. Moreover, the development trend and research focus are given. It is concluded that incremental learning is still a hot research area and will be for a long period. More attention should be paid to the exploration of both biological systems and computational models. [ABSTRACT FROM AUTHOR]

Regularization of neural networks can alleviate overfitting in the training phase. Current regularizationmethods, such as Dropout and DropConnect, randomly drop neural nodes or connections based on a uniform prior. Such a data-independent strategy does not take into consideration of the quality of individual unit or connection. In this paper, we aim to develop a data-dependent approach to regularizing neural network in the framework of Information Geometry. A measurement for the quality of connections is proposed, namely confidence. Specifically, the confidence of a connection is derived from its contribution to the Fisher information distance. The network is adjusted by retaining the confident connections and discarding the less confident ones. The adjusted network, named as ConfNet, would carry the majority of variations in the sample data. The relationships among confidence estimation, Maximum Likelihood Estimation and classical model selection criteria (like Akaike information criterion) is investigated and discussed theoretically. Furthermore, a Stochastic ConfNet is designed by adding a self-adaptive probabilistic sampling strategy. The proposed data-dependent regularization methods achieve promising experimental results on three data collections including MNIST, CIFAR-10 and CIFAR-100. [ABSTRACT FROM AUTHOR]

Handwritten signatures are widely utilized as a form of personal recognition. However, they have the unfortunate shortcoming of being easily abused by those who would fake the identification or intent of an individual which might be very harmful. Therefore, the need for an automatic signature recognition system is crucial. In this paper, a signature recognition approach based on a probabilistic neural network (PNN) and wavelet transform average framing entropy (AFE) is proposed. The system was tested with a wavelet packet (WP) entropy denoted as a WP entropy neural network system (WPENN) and with a discrete wavelet transform (DWT) entropy denoted as a DWT entropy neural network system (DWENN). Our investigation was conducted over several wavelet families and different entropy types. Identification tasks, as well as verification tasks, were investigated for a comprehensive signature system study. Several other methods used in the literature were considered for comparison. Two databases were used for algorithm testing. The best recognition rate result was achieved by WPENN whereby the threshold entropy reached 92%. [ABSTRACT FROM AUTHOR]

Face verification for unrestricted faces in the wild is a challenging task. This paper proposes a method based on two deep convolutional neural networks (CNN) for face verification. In this work, we explore using identification signals to supervise one CNN and the combination of semi-verification and identification to train the other one. In order to estimate semi-verification loss at a low computation cost, a circle, which is composed of all faces, is used for selecting face pairs from pairwise samples. In the process of face normalization, we propose using different landmarks of faces to solve the problems caused by poses. In addition, the final face representation is formed by the concatenating feature of each deep CNN after principal component analysis (PCA) reduction. Furthermore, each feature is a combination of multi-scale representations through making use of auxiliary classifiers. For the final verification, we only adopt the face representation of one region and one resolution of a face jointing Joint Bayesian classifier. Experiments show that our method can extract effective face representation with a small training dataset and our algorithm achieves 99.71% verification accuracy on Labeled Faces in theWild (LFW) dataset. [ABSTRACT FROM AUTHOR]

Computed tomography (CT) images play a vital role in diagnosing rib fractures and determining the severity of chest trauma. However, quickly and accurately identifying rib fractures in a large number of CT images is an arduous task for radiologists. We propose a U-net-based detection method designed to extract rib fracture features at the pixel level to find rib fractures rapidly and precisely. Two modules are applied to the segmentation network—a combined attention module (CAM) and a hybrid dense dilated convolution module (HDDC). The features of the same layer of the encoder and the decoder are fused through CAM, strengthening the local features of the subtle fracture area and increasing the edge features. HDDC is used between the encoder and decoder to obtain sufficient semantic information. Experiments show that on the public dataset, the model test brings the effects of Recall (81.71%), F1 (81.86%), and Dice (53.28%). Experienced radiologists reach lower false positives for each scan, whereas they have underperforming neural network models in terms of detection sensitivities with a long time diagnosis. With the aid of our model, radiologists can achieve higher detection sensitivities than computer-only or human-only diagnosis. [ABSTRACT FROM AUTHOR]

Deep learning can be applied in the field of fault diagnosis without an accurate mechanism model. However, the accurate diagnosis of minor faults using deep learning is limited by the training sample size. In the case that only a small number of noise-polluted samples is available, it is crucial to design a new learning mechanism for the training of deep neural networks to make it more powerful in feature representation. The new learning mechanism for deep neural networks model is accomplished by designing a new loss function such that accurate feature representation driven by consistency of trend features and accurate fault classification driven by consistency of fault direction both can be secured. In such a way, a more robust and more reliable fault diagnosis model using deep neural networks can be established to effectively discriminate those faults with equal or similar membership values of fault classifiers, which is unavailable for traditional methods. Validation for gearbox fault diagnosis shows that 100 training samples polluted with strong noise are adequate for the proposed method to successfully train deep neural networks to achieve satisfactory fault diagnosis accuracy, while more than 1500 training samples are required for traditional methods to achieve comparative fault diagnosis accuracy. [ABSTRACT FROM AUTHOR]

Natural language processing (NLP) models based on deep neural networks (DNNs) are vulnerable to backdoor attacks. Existing backdoor defense methods have limited effectiveness and coverage scenarios. We propose a textual backdoor defense method based on deep feature classification. The method includes deep feature extraction and classifier construction. The method exploits the distinguishability of deep features of poisoned data and benign data. Backdoor defense is implemented in both offline and online scenarios. We conducted defense experiments on two datasets and two models for a variety of backdoor attacks. The experimental results demonstrate the effectiveness of this defense approach and outperform the baseline defense method. [ABSTRACT FROM AUTHOR]

Current breakthroughs in the field of machine learning are fueled by the deployment of deep neural network models. Deep neural networks models are notorious for their dependence on large amounts of labeled data for training them. Active learning is being used as a solution to train classification models with less labeled instances by selecting only the most informative instances for labeling. This is especially important when the labeled data are scarce or the labeling process is expensive. In this paper, we study the application of active learning on attributed graphs. In this setting, the data instances are represented as nodes of an attributed graph. Graph neural networks achieve the current state-of-the-art classification performance on attributed graphs. The performance of graph neural networks relies on the careful tuning of their hyperparameters, usually performed using a validation set, an additional set of labeled instances. In label scarce problems, it is realistic to use all labeled instances for training the model. In this setting, we perform a fair comparison of the existing active learning algorithms proposed for graph neural networks as well as other data types such as images and text. With empirical results, we demonstrate that state-of-the-art active learning algorithms designed for other data types do not perform well on graph-structured data. We study the problem within the framework of the exploration-vs.-exploitation trade-off and propose a new count-based exploration term. With empirical evidence on multiple benchmark graphs, we highlight the importance of complementing uncertainty-based active learning models with an exploration term. [ABSTRACT FROM AUTHOR]

To generate more high-quality aerodynamic data using the information provided by different fidelity data, where low-fidelity aerodynamic data provides the trend information and high-fidelity aerodynamic data provides value information, we applied a deep neural network (DNN) algorithm to fuse the information of multi-fidelity aerodynamic data. We discuss the relationships between the low-fidelity and high-fidelity data, and then we describe the proposed architecture for an aerodynamic data fusion model. The architecture consists of three fully-connected neural networks that are employed to approximate low-fidelity data, and the linear part and nonlinear part of correlation for the low- and high-fidelity data, respectively. To test the proposed multi-fidelity aerodynamic data fusion method, we calculated Euler and Navier–Stokes simulations for a typical airfoil at various Mach numbers and angles of attack to obtain the aerodynamic coefficients as low- and high-fidelity data. A fusion model of the longitudinal coefficients of lift C L and drag C D was constructed with the proposed method. For comparisons, variable complexity modeling and cokriging models were also built. The accuracy spread between the predicted value and true value was discussed for both the training and test data of the three different methods. We calculated the root mean square error and average relative deviation to demonstrate the performance of the three different methods. The fusion result of the proposed method was satisfactory on the test case, and showed a better performance compared with the other two traditional methods presented. The results provide evidence that the method proposed in this paper can be useful in dealing with the multi-fidelity aerodynamic data fusion problem. [ABSTRACT FROM AUTHOR]

The application of machine learning methods to particle physics often does not provide enough understanding of the underlying physics. An interpretable model which provides a way to improve our knowledge of the mechanism governing a physical system directly from the data can be very useful. In this paper, we introduce a simple artificial physical generator based on the Quantum chromodynamical (QCD) fragmentation process. The data simulated from the generator are then passed to a neural network model which we base only on the partial knowledge of the generator. We aimed to see if the interpretation of the generated data can provide the probability distributions of basic processes of such a physical system. This way, some of the information we omitted from the network model on purpose is recovered. We believe this approach can be beneficial in the analysis of real QCD processes. [ABSTRACT FROM AUTHOR]

Convolutional neural networks have powerful performances in many visual tasks because of their hierarchical structures and powerful feature extraction capabilities. SPD (symmetric positive definition) matrix is paid attention to in visual classification, because it has excellent ability to learn proper statistical representation and distinguish samples with different information. In this paper, a deep neural network signal detection method based on spectral convolution features is proposed. In this method, local features extracted from convolutional neural network are used to construct the SPD matrix, and a deep learning algorithm for the SPD matrix is used to detect target signals. Feature maps extracted by two kinds of convolutional neural network models are applied in this study. Based on this method, signal detection has become a binary classification problem of signals in samples. In order to prove the availability and superiority of this method, simulated and semi-physical simulated data sets are used. The results show that, under low SCR (signal-to-clutter ratio), compared with the spectral signal detection method based on the deep neural network, this method can obtain a gain of 0.5–2 dB on simulated data sets and semi-physical simulated data sets. [ABSTRACT FROM AUTHOR]

Determining the coupling between systems remains a topic of active research in the field of complex science. Identifying the proper causal influences in time series can already be very challenging in the trivariate case, particularly when the interactions are non-linear. In this paper, the coupling between three Lorenz systems is investigated with the help of specifically designed artificial neural networks, called time delay neural networks (TDNNs). TDNNs can learn from their previous inputs and are therefore well suited to extract the causal relationship between time series. The performances of the TDNNs tested have always been very positive, showing an excellent capability to identify the correct causal relationships in absence of significant noise. The first tests on the time localization of the mutual influences and the effects of Gaussian noise have also provided very encouraging results. Even if further assessments are necessary, the networks of the proposed architecture have the potential to be a good complement to the other techniques available in the market for the investigation of mutual influences between time series. [ABSTRACT FROM AUTHOR]

This paper proposes a deep convolutional neural network model with encoder-decoder architecture to extract road network from satellite images. We employ ResNet-18 and Atrous Spatial Pyramid Pooling technique to trade off between the extraction precision and running time. A modified cross entropy loss function is proposed to train our deep model. A PointRend algorithm is used to recover a smooth, clear and sharp road boundary. The augmentated DeepGlobe dataset is used to train our deep model and the asynchronous training method is applied to accelerate the training process. Five salellite images covering Xiaomu village are taken as input to evaluate our model. The proposed E-Road model has fewer number of parameters and shorter training time. The experiments show E-Road outperforms other state-of-the-art deep models with 5.84% to 59.09% improvement, and can give the accurate predictions for the images with complex environment. [ABSTRACT FROM AUTHOR]

This paper analyzes and compares the sensitivity and suitability of several artificial intelligence techniques applied to the Morgan–Keenan (MK) system for the classification of stars. The MK system is based on a sequence of spectral prototypes that allows classifying stars according to their effective temperature and luminosity through the study of their optical stellar spectra. Here, we include the method description and the results achieved by the different intelligent models developed thus far in our ongoing stellar classification project: fuzzy knowledge-based systems, backpropagation, radial basis function (RBF) and Kohonen artificial neural networks. Since one of today's major challenges in this area of astrophysics is the exploitation of large terrestrial and space databases, we propose a final hybrid system that integrates the best intelligent techniques, automatically collects the most important spectral features, and determines the spectral type and luminosity level of the stars according to the MK standard system. This hybrid approach truly emulates the behavior of human experts in this area, resulting in higher success rates than any of the individual implemented techniques. In the final classification system, the most suitable methods are selected for each individual spectrum, which implies a remarkable contribution to the automatic classification process. [ABSTRACT FROM AUTHOR]

In this paper, a new class of impulsive neural networks with fractional-like derivatives is defined, and the practical stability properties of the solutions are investigated. The stability analysis exploits a new type of Lyapunov-like functions and their derivatives. Furthermore, the obtained results are applied to a bidirectional associative memory (BAM) neural network model with fractional-like derivatives. Some new results for the introduced neural network models with uncertain values of the parameters are also obtained. [ABSTRACT FROM AUTHOR]

Decoding motor imagery (MI) electroencephalogram (EEG) signals for brain-computer interfaces (BCIs) is a challenging task because of the severe non-stationarity of perceptual decision processes. Recently, deep learning techniques have had great success in EEG decoding because of their prominent ability to learn features from raw EEG signals automatically. However, the challenge that the deep learning method faces is that the shortage of labeled EEG signals and EEGs sampled from other subjects cannot be used directly to train a convolutional neural network (ConvNet) for a target subject. To solve this problem, in this paper, we present a novel conditional domain adaptation neural network (CDAN) framework for MI EEG signal decoding. Specifically, in the CDAN, a densely connected ConvNet is firstly applied to obtain high-level discriminative features from raw EEG time series. Then, a novel conditional domain discriminator is introduced to work as an adversarial with the label classifier to learn commonly shared intra-subjects EEG features. As a result, the CDAN model trained with sufficient EEG signals from other subjects can be used to classify the signals from the target subject efficiently. Competitive experimental results on a public EEG dataset (High Gamma Dataset) against the state-of-the-art methods demonstrate the efficacy of the proposed framework in recognizing MI EEG signals, indicating its effectiveness in automatic perceptual decision decoding. [ABSTRACT FROM AUTHOR]

Personnel performance is important for the high-technology industry to ensure its core competitive advantages are present. Therefore, predicting personnel performance is an important research area in human resource management (HRM). In this paper, to improve prediction performance, we propose a novel framework for personnel performance prediction to help decision-makers to forecast future personnel performance and recruit the best suitable talents. Firstly, a hybrid convolutional recurrent neural network (CRNN) model based on self-attention mechanism is presented, which can automatically learn discriminative features and capture global contextual information from personnel performance data. Moreover, we treat the prediction problem as a classification task. Then, the k-nearest neighbor (KNN) classifier was used to predict personnel performance. The proposed framework is applied to a real case of personnel performance prediction. The experimental results demonstrate that the presented approach achieves significant performance improvement for personnel performance compared to existing methods. [ABSTRACT FROM AUTHOR]

The motor imagery-based brain-computer interface (BCI) using electroencephalography (EEG) has been receiving attention from neural engineering researchers and is being applied to various rehabilitation applications. However, the performance degradation caused by motor imagery EEG with very low single-to-noise ratio faces several application issues with the use of a BCI system. In this paper, we propose a novel motor imagery classification scheme based on the continuous wavelet transform and the convolutional neural network. Continuous wavelet transform with three mother wavelets is used to capture a highly informative EEG image by combining time-frequency and electrode location. A convolutional neural network is then designed to both classify motor imagery tasks and reduce computation complexity. The proposed method was validated using two public BCI datasets, BCI competition IV dataset 2b and BCI competition II dataset III. The proposed methods were found to achieve improved classification performance compared with the existing methods, thus showcasing the feasibility of motor imagery BCI. [ABSTRACT FROM AUTHOR]

Recently, convolutional neural network (CNN) based on the encoder-decoder structure have been successfully applied to image dehazing. However, these CNN based dehazing methods have two limitations: First, these dehazing models are large in size with enormous parameters, which not only consumes much GPU memory, but also is hard to train from scratch. Second, these models, which ignore the structural information at different resolutions of intermediate layers, cannot capture informative texture and edge information for dehazing by stacking more layers. In this paper, we propose a light-weight end-to-end network named the residual dense pyramid network (RDPN) to address the above problems. To exploit the structural information at different resolutions of intermediate layers fully, a new residual dense pyramid (RDP) is proposed as a building block. By introducing a dense information fusion layer and the residual learning module, the RDP can maximize the information flow and extract local features. Furthermore, the RDP further learns the structural information from intermediate layers via a multiscale pyramid fusion mechanism. To reduce the number of network parameters and to ease the training process, we use one RDP in the encoder and two RDPs in the decoder, following a multilevel pyramid pooling layer for incorporating global context features before estimating the final result. The extensive experimental results on a synthetic dataset and real-world images demonstrate that the new RDPN achieves favourable performance compared with some state-of-the-art methods, e.g., the recent densely connected pyramid dehazing network, the all-in-one dehazing network, the enhanced pix2pix dehazing network, pixel-based alpha blending, artificial multi-exposure image fusions and the genetic programming estimator, in terms of accuracy, run time and number of parameters. To be specific, RDPN outperforms all of the above methods in terms of PSNR by at least 4.25 dB. The run time of the proposed method is 0.021 s, and the number of parameters is 1,534,799, only 6% of that used by the densely connected pyramid dehazing network. [ABSTRACT FROM AUTHOR]

Image retrieval based on a convolutional neural network (CNN) has attracted great attention among researchers because of the high performance. The pooling method has become a research hotpot in the task of image retrieval in recent years. In this paper, we propose the feature distribution entropy (FDE) to measure the difference of regional distribution information in the feature maps from CNNs. We propose a novel pooling method, which fuses our proposed FDE with region maximum activations of convolutions (R-MAC) features to improve the performance of image retrieval, as it takes the advantage of regional distribution information in the feature maps. Compared with the descriptors computed by R-MAC pooling, our proposed method considers not only the most significant feature values of each region in feature map, but also the distribution difference in different regions. We utilize the histogram of feature values to calculate regional distribution entropy and concatenate the regional distribution entropy into FDE, which is further normalized and fused with R-MAC feature vectors by weighted summation to generate the final feature descriptors. We have conducted experiments on public datasets and the results demonstrate that our proposed method could produce better retrieval performances than existing state-of-the-art algorithms. Further, higher performance could be achieved by performing these post-processing on the improved feature descriptors. [ABSTRACT FROM AUTHOR]

In this paper, we study the statistical properties of the stationary firing-rate states of a neural network model with quenched disorder. The model has arbitrary size, discrete-time evolution equations and binary firing rates, while the topology and the strength of the synaptic connections are randomly generated from known, generally arbitrary, probability distributions. We derived semi-analytical expressions of the occurrence probability of the stationary states and the mean multistability diagram of the model, in terms of the distribution of the synaptic connections and of the external stimuli to the network. Our calculations rely on the probability distribution of the bifurcation points of the stationary states with respect to the external stimuli, calculated in terms of the permanent of special matrices using extreme value theory. While our semi-analytical expressions are exact for any size of the network and for any distribution of the synaptic connections, we focus our study on networks made of several populations, that we term "statistically homogeneous" to indicate that the probability distribution of their connections depends only on the pre- and post-synaptic population indexes, and not on the individual synaptic pair indexes. In this specific case, we calculated analytically the permanent, obtaining a compact formula that outperforms of several orders of magnitude the Balasubramanian-Bax-Franklin-Glynn algorithm. To conclude, by applying the Fisher-Tippett-Gnedenko theorem, we derived asymptotic expressions of the stationary-state statistics of multi-population networks in the large-network-size limit, in terms of the Gumbel (double exponential) distribution. We also provide a Python implementation of our formulas and some examples of the results generated by the code. [ABSTRACT FROM AUTHOR]

Automatic coal-rock recognition is one of the critical technologies for intelligent coal mining and processing. Most existing coal-rock recognition methods have some defects, such as unsatisfactory performance and low robustness. To solve these problems, and taking distinctive visual features of coal and rock into consideration, the multi-scale feature fusion coal-rock recognition (MFFCRR) model based on a multi-scale Completed Local Binary Pattern (CLBP) and a Convolution Neural Network (CNN) is proposed in this paper. Firstly, the multi-scale CLBP features are extracted from coal-rock image samples in the Texture Feature Extraction (TFE) sub-model, which represents texture information of the coal-rock image. Secondly, the high-level deep features are extracted from coal-rock image samples in the Deep Feature Extraction (DFE) sub-model, which represents macroscopic information of the coal-rock image. The texture information and macroscopic information are acquired based on information theory. Thirdly, the multi-scale feature vector is generated by fusing the multi-scale CLBP feature vector and deep feature vector. Finally, multi-scale feature vectors are input to the nearest neighbor classifier with the chi-square distance to realize coal-rock recognition. Experimental results show the coal-rock image recognition accuracy of the proposed MFFCRR model reaches 97.9167%, which increased by 2%–3% compared with state-of-the-art coal-rock recognition methods. [ABSTRACT FROM AUTHOR]

Face recognition is a popular and efficient form of biometric authentication used in many software applications. One drawback of this technique is that it is prone to face spoofing attacks, where an impostor can gain access to the system by presenting a photograph of a valid user to the sensor. Thus, face liveness detection is a necessary step before granting authentication to the user. In this paper, we have developed deep architectures for face liveness detection that use a combination of texture analysis and a convolutional neural network (CNN) to classify the captured image as real or fake. Our development greatly improved upon a recent approach that applies nonlinear diffusion based on an additive operator splitting scheme and a tridiagonal matrix block-solver algorithm to the image, which enhances the edges and surface texture in the real image. We then fed the diffused image to a deep CNN to identify the complex and deep features for classification. We obtained 100% accuracy on the NUAA Photograph Impostor dataset for face liveness detection using one of our enhanced architectures. Further, we gained insight into the enhancement of the face liveness detection architecture by evaluating three different deep architectures, which included deep CNN, residual network, and the inception network version 4. We evaluated the performance of each of these architectures on the NUAA dataset and present here the experimental results showing under what conditions an architecture would be better suited for face liveness detection. While the residual network gave us competitive results, the inception network version 4 produced the optimal accuracy of 100% in liveness detection (with nonlinear anisotropic diffused images with a smoothness parameter of 15). Our approach outperformed all current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In a decentralized multi-robot exploration problem, the robots have to cooperate effectively to map a strange environment as soon as possible without a centralized controller. In the past few decades, a set of "human-designed" cooperation strategies have been proposed to address this problem, such as the well-known frontier-based approach. However, many real-world settings, especially the ones that are constantly changing, are too complex for humans to design efficient and decentralized strategies. This paper presents a novel approach, the Attention-based Communication neural network (CommAttn), to "learn" the cooperation strategies automatically in the decentralized multi-robot exploration problem. The communication neural network enables the robots to learn the cooperation strategies with explicit communication. Moreover, the attention mechanism we introduced additionally can precisely calculate whether the communication is necessary for each pair of agents by considering the relevance of each received message, which enables the robots to communicate only with the necessary partners. The empirical results on a simulated multi-robot disaster exploration scenario demonstrate that our proposal outperforms the traditional "human-designed" methods, as well as other competing "learning-based" methods in the exploration task. [ABSTRACT FROM AUTHOR]

Although Shannon mutual information has been widely used, its effective calculation is often difficult for many practical problems, including those in neural population coding. Asymptotic formulas based on Fisher information sometimes provide accurate approximations to the mutual information but this approach is restricted to continuous variables because the calculation of Fisher information requires derivatives with respect to the encoded variables. In this paper, we consider information-theoretic bounds and approximations of the mutual information based on Kullback-Leibler divergence and Rényi divergence. We propose several information metrics to approximate Shannon mutual information in the context of neural population coding. While our asymptotic formulas all work for discrete variables, one of them has consistent performance and high accuracy regardless of whether the encoded variables are discrete or continuous. We performed numerical simulations and confirmed that our approximation formulas were highly accurate for approximating the mutual information between the stimuli and the responses of a large neural population. These approximation formulas may potentially bring convenience to the applications of information theory to many practical and theoretical problems. [ABSTRACT FROM AUTHOR]

In this paper, a new three-dimensional fractional-order Hopfield-type neural network with delay is proposed. The system has a unique equilibrium point at the origin, which is a saddle point with index two, hence unstable. Intermittent chaos is found in this system. The complex dynamics are analyzed both theoretically and numerically, including intermittent chaos, periodicity, and stability. Those phenomena are confirmed by phase portraits, bifurcation diagrams, and the Largest Lyapunov exponent. Furthermore, a synchronization method based on the state observer is proposed to synchronize a class of time-delayed fractional-order Hopfield-type neural networks. [ABSTRACT FROM AUTHOR]

Detecting and classifying ships based on radiated noise provide practical guidelines for the reduction of underwater noise footprint of shipping. In this paper, the detection and classification are implemented by auditory inspired convolutional neural networks trained from raw underwater acoustic signal. The proposed model includes three parts. The first part is performed by a multi-scale 1D time convolutional layer initialized by auditory filter banks. Signals are decomposed into frequency components by convolution operation. In the second part, the decomposed signals are converted into frequency domain by permute layer and energy pooling layer to form frequency distribution in auditory cortex. Then, 2D frequency convolutional layers are applied to discover spectro-temporal patterns, as well as preserve locality and reduce spectral variations in ship noise. In the third part, the whole model is optimized with an objective function of classification to obtain appropriate auditory filters and feature representations that are correlative with ship categories. The optimization reflects the plasticity of auditory system. Experiments on five ship types and background noise show that the proposed approach achieved an overall classification accuracy of 79.2%, which improved by 6% compared to conventional approaches. Auditory filter banks were adaptive in shape to improve accuracy of classification. [ABSTRACT FROM AUTHOR]

In this paper, a fault-tolerant control scheme is presented for a class of stochastic distribution collaborative control systems, which are composed of three subsystems connected in series to complete the control target. The radial basis function neural network is used to approximate the output probability density function of the third subsystem, which is also the output of the entire system. When fault occurs in the first subsystem, an adaptive diagnostic observer is designed to estimate the value of fault. However, the first subsystem does not have the ability of self-recovery, minimum rational entropy controllers are designed in the latter subsystems to compensate the influence of the fault and minimize the entropy of the system output. A numerical simulation is given to verify the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

In this paper, a deep learning approach, Restricted Boltzmann Machine (RBM), is used to perform automatic hand sign language recognition from visual data. We evaluate how RBM, as a deep generative model, is capable of generating the distribution of the input data for an enhanced recognition of unseen data. Two modalities, RGB and Depth, are considered in the model input in three forms: original image, cropped image, and noisy cropped image. Five crops of the input image are used and the hand of these cropped images are detected using Convolutional Neural Network (CNN). After that, three types of the detected hand images are generated for each modality and input to RBMs. The outputs of the RBMs for two modalities are fused in another RBM in order to recognize the output sign label of the input image. The proposed multi-modal model is trained on all and part of the American alphabet and digits of four publicly available datasets. We also evaluate the robustness of the proposal against noise. Experimental results show that the proposed multi-modal model, using crops and the RBM fusing methodology, achieves state-of-the-art results on Massey University Gesture Dataset 2012, American Sign Language (ASL). and Fingerspelling Dataset from the University of Surrey's Center for Vision, Speech and Signal Processing, NYU, and ASL Fingerspelling A datasets. [ABSTRACT FROM AUTHOR]

Much research on adversarial attacks has proved that deep neural networks have certain security vulnerabilities. Among potential attacks, black-box adversarial attacks are considered the most realistic based on the the natural hidden nature of deep neural networks. Such attacks have become a critical academic emphasis in the current security field. However, current black-box attack methods still have shortcomings, resulting in incomplete utilization of query information. Our research, based on the newly proposed Simulator Attack, proves the correctness and usability of feature layer information in a simulator model obtained by meta-learning for the first time. Then, we propose an optimized Simulator Attack+ based on this discovery. Our optimization methods used in Simulator Attack+ include: (1) a feature attentional boosting module that uses the feature layer information of the simulator to enhance the attack and accelerate the generation of adversarial examples; (2) a linear self-adaptive simulator-predict interval mechanism that allows the simulator model to be fully fine-tuned in the early stage of the attack and dynamically adjusts the interval for querying the black-box model; and (3) an unsupervised clustering module to provide a warm-start for targeted attacks. Results from experiments on the CIFAR-10 and CIFAR-100 datasets clearly show that Simulator Attack+ can further reduce the number of consuming queries to improve query efficiency while maintaining the attack. [ABSTRACT FROM AUTHOR]

Shadow is one of the fundamental indicators of remote sensing image which could cause loss or interference of the target data. As a result, the detection and removal of shadow has already been the hotspot of current study because of the complicated background information. In the following passage, a model combining the Atmospheric Transport Model (hereinafter abbreviated as ATM) with the Poisson Equation, AP ShadowNet, is proposed for the shadow detection and removal of remote sensing images by unsupervised learning. This network based on a preprocessing network based on ATM, A Net, and a network based on the Poisson Equation, P Net. Firstly, corresponding mapping between shadow and unshaded area is generated by the ATM. The brightened image will then enter the Confrontation identification in the P Net. Lastly, the reconstructed image is optimized on color consistency and edge transition by Poisson Equation. At present, most shadow removal models based on neural networks are significantly data-driven. Fortunately, by the model in this passage, the unsupervised shadow detection and removal could be released from the data source restrictions from the remote sensing images themselves. By verifying the shadow removal on our model, the result shows a satisfying effect from a both qualitative and quantitative angle. From a qualitative point of view, our results have a prominent effect on tone consistency and removal of detailed shadows. From the quantitative point of view, we adopt the non-reference evaluation indicators: gradient structure similarity (NRSS) and Natural Image Quality Evaluator (NIQE). Combining various evaluation factors such as reasoning speed and memory occupation, it shows that it is outstanding among other current algorithms. [ABSTRACT FROM AUTHOR]

At present, the success of most intelligent fault diagnosis methods is heavily dependent on large datasets of artificial simulation faults (ASF), which have not been widely used in practice because it is often costly to obtain a large number of samples in reality. Fortunately, various faults can be easily simulated in the laboratory, and these simulated faults contain a lot of fault diagnosis knowledge. In this study, based on a Siamese network framework, we propose a bearing fault diagnosis based on few-shot transfer learning across different datasets (cross-machine), using the knowledge of ASF to diagnose bearings with natural faults (NF). First of all, the model obtains a good feature encoder in the source domain, then defines a fault support set for comparison, and finally adjusts the support set with a very small number of target domain samples to improve the fault diagnosis performance of the model. We carried out experimental verification from many aspects on the ASF and NF datasets provided by Case Western Reserve University (CWRU) and Paderborn University (PU). The results show that the proposed method can fully learn diagnostic knowledge in different ASF datasets and sample numbers, and effectively use this knowledge to accurately identify the health state of the NF bearing, which has strong generalization and robustness. Our method does not need second training, which may be more convenient in some practical applications. Finally, we also discuss the possible limitations of this method. [ABSTRACT FROM AUTHOR]

The calculation of the Augmented Inverse Probability Weighting (AIPW) estimator of the Average Treatment Effect (ATE) is carried out in two steps, where in the first step, the treatment and outcome are modeled, and in the second step, the predictions are inserted into the AIPW estimator. The model misspecification in the first step has led researchers to utilize Machine Learning algorithms instead of parametric algorithms. However, the existence of strong confounders and/or Instrumental Variables (IVs) can lead the complex ML algorithms to provide perfect predictions for the treatment model which can violate the positivity assumption and elevate the variance of AIPW estimators. Thus the complexity of ML algorithms must be controlled to avoid perfect predictions for the treatment model while still learning the relationship between the confounders and the treatment and outcome. We use two NN architectures with an L 1 -regularization on specific NN parameters and investigate how their certain hyperparameters should be tuned in the presence of confounders and IVs to achieve a low bias-variance tradeoff for ATE estimators such as AIPW estimator. Through simulation results, we will provide recommendations as to how NNs can be employed for ATE estimation. [ABSTRACT FROM AUTHOR]

The present study concerns the modeling of the thermal behavior of a porous longitudinal fin under fully wetted conditions with linear, quadratic, and exponential thermal conductivities surrounded by environments that are convective, conductive, and radiative. Porous fins are widely used in various engineering and everyday life applications. The Darcy model was used to formulate the governing non-linear singular differential equation for the heat transfer phenomenon in the fin. The universal approximation power of multilayer perceptron artificial neural networks (ANN) was applied to establish a model of approximate solutions for the singular non-linear boundary value problem. The optimization strategy of a sports-inspired meta-heuristic paradigm, the Tiki-Taka algorithm (TTA) with sequential quadratic programming (SQP), was utilized to determine the thermal performance and the effective use of fins for diverse values of physical parameters, such as parameter for the moist porous medium, dimensionless ambient temperature, radiation coefficient, power index, in-homogeneity index, convection coefficient, and dimensionless temperature. The results of the designed ANN-TTA-SQP algorithm were validated by comparison with state-of-the-art techniques, including the whale optimization algorithm (WOA), cuckoo search algorithm (CSA), grey wolf optimization (GWO) algorithm, particle swarm optimization (PSO) algorithm, and machine learning algorithms. The percentage of absolute errors and the mean square error in the solutions of the proposed technique were found to lie between 10 − 4 to 10 − 5 and 10 − 8 to 10 − 10 , respectively. A comprehensive study of graphs, statistics of the solutions, and errors demonstrated that the proposed scheme's results were accurate, stable, and reliable. It was concluded that the pace at which heat is transferred from the surface of the fin to the surrounding environment increases in proportion to the degree to which the wet porosity parameter is increased. At the same time, inverse behavior was observed for increase in the power index. The results obtained may support the structural design of thermally effective cooling methods for various electronic consumer devices. [ABSTRACT FROM AUTHOR]

With the successful development in computer vision, building a deep convolutional neural network (CNNs) has been mainstream, considering the character of shared parameters in a convolutional layer. Stacking convolutional layers into a deep structure improves performance, but over-stacking also ramps up the needed resources for GPUs. Seeing another surge of Transformers in computer vision, the issue has aroused severely. A resource-hungry model is hardly implemented for limited hardware or single-customers-based GPU. Therefore, this work focuses on these concerns and proposes an efficient but robust backbone, which equips with channel and spatial direction attentions, so the attentions help to expand receptive fields in shallow convolutional layers and pass the information to every layer. An attention-boosted network based on already efficient CNNs, Universal Pixel Attention Networks (UPANets), is proposed. Through a series of experiments, UPANets fulfil the purposes of learning global information with less needed resources and outshine many existing SOTAs in CIFAR-{10, 100}. [ABSTRACT FROM AUTHOR]

Resolution is an intuitive assessment for the visual quality of images, which is limited by physical devices. Recently, image super-resolution (SR) models based on deep convolutional neural networks (CNNs) have made significant progress. However, most existing SR models require high computational costs with network depth, hindering practical application. In addition, these models treat intermediate features equally and rarely explore the discriminative capacity hidden in their abundant features. To tackle these issues, we propose an attention network with information distillation(AIDN) for efficient and accurate image super-resolution, which adaptively modulates the feature responses by modeling the interactions between channel dimension and spatial features. Specifically, gated channel transformation (GCT) is introduced to gather global contextual information among different channels to modulate intermediate high-level features. Moreover, a recalibrated attention module (RAM) is proposed to rescale these feature responses, and RAM concentrates the essential contents around spatial locations. Benefiting from the gated channel transformation and spatial information masks working jointly, our proposed AIDN can obtain a more powerful ability to identify information. It effectively improves computational efficiency while improving reconstruction accuracy. Comprehensive quantitative and qualitative evaluations demonstrate that our AIDN outperforms state-of-the-art models in terms of reconstruction performance and visual quality. [ABSTRACT FROM AUTHOR]

Graph neural networks (GNNs) with feature propagation have demonstrated their power in handling unstructured data. However, feature propagation is also a smooth process that tends to make all node representations similar as the number of propagation increases. To address this problem, we propose a novel Block-Based Adaptive Decoupling (BBAD) Framework to produce effective deep GNNs by utilizing backbone networks. In this framework, each block contains a shallow GNN with feature propagation and transformation decoupled. We also introduce layer regularizations and flexible receptive fields to automatically adjust the propagation depth and to provide different aggregation hops for each node, respectively. We prove that the traditional coupled GNNs are more likely to suffer from over-smoothing when they become deep. We also demonstrate the diversity of outputs from different blocks of our framework. In the experiments, we conduct semi-supervised and fully supervised node classifications on benchmark datasets, and the results verify that our method can not only improve the performance of various backbone networks, but also is superior to existing deep graph neural networks with less parameters. [ABSTRACT FROM AUTHOR]

We show that neural networks with an absolute value activation function and with network path norm, network sizes and network weights having logarithmic dependence on 1 / ε can ε -approximate functions that are analytic on certain regions of C d . [ABSTRACT FROM AUTHOR]

One of the major requirements of content based image retrieval (CBIR) systems is to ensure meaningful image retrieval against query images. The performance of these systems is severely degraded by the inclusion of image content which does not contain the objects of interest in an image during the image representation phase. Segmentation of the images is considered as a solution but there is no technique that can guarantee the object extraction in a robust way. Another limitation of the segmentation is that most of the image segmentation techniques are slow and their results are not reliable. To overcome these problems, a bandelet transform based image representation technique is presented in this paper, which reliably returns the information about the major objects found in an image. For image retrieval purposes, artificial neural networks (ANN) are applied and the performance of the system and achievement is evaluated on three standard data sets used in the domain of CBIR. [ABSTRACT FROM AUTHOR]

Spiking neural P systems (SN P systems for short) realize the high abstraction and simulation of the working mechanism of the human brain, and adopts spikes for information encoding and processing, which are regarded as one of the third-generation neural network models. In the nervous system, the conduction of excitation depends on the presence of membrane potential (also known as the transmembrane potential difference), and the conduction of excitation on neurons is the conduction of action potentials. On the basis of the SN P systems with polarizations, in which the neuron-associated polarization is the trigger condition of the rule, the concept of neuronal membrane potential is introduced into systems. The obtained variant of the SN P system features charge accumulation and computation within neurons in quantity, as well as transmission between neurons. In addition, there are inhibitory synapses between neurons that inhibit excitatory transmission, and as such, synapses cause postsynaptic neurons to generate inhibitory postsynaptic potentials. Therefore, to make the model better fit the biological facts, inhibitory rules and anti-spikes are also adopted to obtain the spiking neural P systems with membrane potentials, inhibitory rules, and anti-spikes (referred to as the MPAIRSN P systems). The Turing universality of the MPAIRSN P systems as number generating and accepting devices is demonstrated. On the basis of the above working mechanism of the system, a small universal MPAIRSN P system with 95 neurons for computing functions is designed. The comparisons with other SN P models conclude that fewer neurons are required by the MPAIRSN P systems to realize universality. [ABSTRACT FROM AUTHOR]

Purpose: In this work, we propose an implementation of the Bienenstock–Cooper–Munro (BCM) model, obtained by a combination of the classical framework and modern deep learning methodologies. The BCM model remains one of the most promising approaches to modeling the synaptic plasticity of neurons, but its application has remained mainly confined to neuroscience simulations and few applications in data science. Methods: To improve the convergence efficiency of the BCM model, we combine the original plasticity rule with the optimization tools of modern deep learning. By numerical simulation on standard benchmark datasets, we prove the efficiency of the BCM model in learning, memorization capacity, and feature extraction. Results: In all the numerical simulations, the visualization of neuronal synaptic weights confirms the memorization of human-interpretable subsets of patterns. We numerically prove that the selectivity obtained by BCM neurons is indicative of an internal feature extraction procedure, useful for patterns clustering and classification. The introduction of competitiveness between neurons in the same BCM network allows the network to modulate the memorization capacity of the model and the consequent model selectivity. Conclusions: The proposed improvements make the BCM model a suitable alternative to standard machine learning techniques for both feature selection and classification tasks. [ABSTRACT FROM AUTHOR]

In this paper, we study Tsallis' fractional entropy (TFE) in a complex domain by applying the definition of the complex probability functions. We study the upper and lower bounds of TFE based on some special functions. Moreover, applications in complex neural networks (CNNs) are illustrated to recognize the accuracy of CNNs. [ABSTRACT FROM AUTHOR]

The sustainability of regional water resources has important supporting data needed for establishing policies on the sustainable development of the social economy. The purpose of this paper is to propose an assessment method to accurately reflect the sustainability of regional water resources in various areas. The method is based on the relative entropy of the information entropy theory. The steps are as follows. Firstly, the pretreatment of the evaluation sample data is required, before the relative entropy of each standard evaluation sample and evaluation grade (SEG) is calculated to obtain the entropy weight of each evaluation index. After this, the entropy weighted comprehensive index (WCI) of the standard evaluation grade sample is obtained. The function relation between WCI and SEG can be fitted by the cubic polynomial to construct the evaluation function. Using the above steps, a generalized entropy method (GEM) for the sustainable assessment of regional water resources is established and it is used to evaluate the sustainability of water resources in the Pingba and Huai River areas in China. The results show that the proposed GEM model can accurately reflect the sustainable water resources in the two regions. Compared with the other evaluation models, such as the Shepherd method, Artificial Neural Network and Fuzzy comprehensive evaluation, the GEM model has larger differences in its evaluation results, which are more reasonable. Thus, the proposed GEM model can provide scientific data support for coordinating the relationship between the sustainable development and utilization of regional water resources in order to improve the development of regional population, society and economy. [ABSTRACT FROM AUTHOR]

Taking inspiration from biology to solve engineering problems using the organizing principles of biological neural computation is the aim of the field of neuromorphic engineering. This field has demonstrated success in sensor based applications (vision and audition) as well as in cognition and actuators. This paper is focused on mimicking the approaching detection functionality of the retina that is computed by one type of Retinal Ganglion Cell (RGC) and its application to robotics. These RGCs transmit action potentials when an expanding object is detected. In this work we compare the software and hardware logic FPGA implementations of this approaching function and the hardware latency when applied to robots, as an attention/reaction mechanism. The visual input for these cells comes from an asynchronous event-driven Dynamic Vision Sensor, which leads to an end-to-end event based processing system. The software model has been developed in Java, and computed with an average processing time per event of 370 ns on a NUC embedded computer. The output firing rate for an approaching object depends on the cell parameters that represent the needed number of input events to reach the firing threshold. For the hardware implementation, on a Spartan 6 FPGA, the processing time is reduced to 160 ns/event with the clock running at 50 MHz. The entropy has been calculated to demonstrate that the system is not totally deterministic in response to approaching objects because of several bioinspired characteristics. It has been measured that a Summit XL mobile robot can react to an approaching object in 90 ms, which can be used as an attentional mechanism. This is faster than similar event-based approaches in robotics and equivalent to human reaction latencies to visual stimulus. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks (ConvNets), which are at the heart of many new emerging applications, achieve remarkable performance in audio and visual recognition tasks. Unfortunately, achieving accuracy often implies significant computational costs, limiting deployability. In modern ConvNets it is typical for the convolution layers to consume the vast majority of computational resources during inference. This has made the acceleration of these layers an important research area in academia and industry. In this paper, we examine the effects of co-optimizing the internal structures of the convolutional layers and underlying implementation of fundamental convolution operation. We demonstrate that a combination of these methods can have a big impact on the overall speedup of a ConvNet, achieving a ten-fold increase over baseline. We also introduce a new class of fast one-dimensional (1D) convolutions for ConvNets using the Toom–Cook algorithm. We show that our proposed scheme is mathematically well-grounded, robust, and does not require any time-consuming retraining, while still achieving speedups solely from convolutional layers with no loss in baseline accuracy. [ABSTRACT FROM AUTHOR]

The convolution neural network (CNN) has achieved state-of-the-art performance in many computer vision applications e.g., classification, recognition, detection, etc. However, the global optimization of CNN training is still a problem. Fast classification and training play a key role in the development of the CNN. We hypothesize that the smoother and optimized the training of a CNN goes, the more efficient the end result becomes. Therefore, in this paper, we implement a modified resilient backpropagation (MRPROP) algorithm to improve the convergence and efficiency of CNN training. Particularly, a tolerant band is introduced to avoid network overtraining, which is incorporated with the global best concept for weight updating criteria to allow the training algorithm of the CNN to optimize its weights more swiftly and precisely. For comparison, we present and analyze four different training algorithms for CNN along with MRPROP, i.e., resilient backpropagation (RPROP), Levenberg–Marquardt (LM), conjugate gradient (CG), and gradient descent with momentum (GDM). Experimental results showcase the merit of the proposed approach on a public face and skin dataset. [ABSTRACT FROM AUTHOR]