Search

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

Blade icing seriously affects wind turbines' aerodynamic performance and output power. Timely and accurately predicting blade icing status is crucial to improving the economy and safety of wind farms. However, existing blade icing prediction methods cannot effectively solve the problems of unbalanced icing/non-icing data and low prediction accuracy. In order to solve the above problems, this paper proposes a wind turbine blade icing prediction method based on the focal loss function and CNN-Attention-GRU. First, the recursive feature elimination method combined with the physical mechanism of icing is used to extract features highly correlated with blade icing, and a new feature subset is formed through a sliding window algorithm. Then, the focal loss function is utilized to assign more weight to the ice samples with a lower proportion, addressing the significant class imbalance between the ice and non-ice categories. Finally, based on the CNN-Attention-GRU algorithm, a blade icing prediction model is established using continuous 24-h historical data as the input and the icing status of the next 24 h as the output. The model is compared with advanced neural network models. The results show that the proposed method improves the prediction accuracy and F1 score by an average of 6.41% and 4.27%, respectively, demonstrating the accuracy and effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The paper presents the application of the artificial neural network algorithm for the retrieval of total precipitable water in the atmosphere over water and land from the measurements of MTVZA-GYa microwave radiometer on board the Meteor-M No. 2-2 satellite. Satellite-based estimates of total precipitable water were compared with radiosonde and AERONET data, as well as with the ECMWF numerical weather prediction model output. According to the comparison, the root-mean-square error (RMSE) does not exceed 4.5 mm for radiosonde data and is less than 4 mm for the ECMWF and AERONET data. The best accuracy is provided over water with the RMSE not exceeding 3 mm. The total precipitable water estimates retrieved from MTVZA-GYa and NOAA-20/ATMS radiometer data are consistent over water, while the MTVZA-GYa based estimates are more accurate over land. [ABSTRACT FROM AUTHOR]

Past research on the process of extinguishing a fire typically used a traditional linear water jet falling point model and the results ignored external factors, such as environmental conditions and the status of the fire engine, even though the water jet falling point location prediction was often associated with these parameters and showed a nonlinear relationship. This paper constructed a BP (Back Propagation) neural network model. The fire gun nozzle characteristics were included as model inputs, and the water discharge point coordinates were the model outputs; thus, the model could precisely predict the water discharge point with small error and high precision to determine an accurate firing position and allow for the timely adjustment of the spray gun. To improve the slow convergence and local optimality problems of the BP neural network (BPNN), this paper further used a genetic algorithm to optimize the BPNN (GA-BPNN). The BPNN can be used to optimize the weights in the network to train them for global optimization. A genetic algorithm was introduced into the neural network approach, and the water jet landing prediction model was further improved. The simulation results showed that the prediction accuracy of the GA-BP model was better than that of the BPNN alone. The established model can accurately predict the location of the water jet, making the prediction results more useful for firefighters. [ABSTRACT FROM AUTHOR]

In this paper, we suggest basing the development of classification methods on traditional techniques but approximating them in whole or in part with artificial neural networks (ANNs). Compared to a direct ANN approach, the underlying traditional method is often easier to analyse. Classification failures can be better understood and corrected for, while at the same time faster execution can be obtained due to the parallel ANN structure. Furthermore, such a two‐step design philosophy partly eliminates the 'guesswork' associated with the design of ANNs. The expected gain is thus that the benefits from the traditional field, and the ANN field can be obtained by combining the best features from both fields. We illustrate our approach by working through an explicit example, namely, a Nearest Neighbour classifier applied to a subset of the MNIST database of handwritten digits. Two different approaches are discussed for how to translate the traditional method into an ANN. The first approach is based on a constructive implementation which directly reflects the original algorithm. The second approach uses ANNs to approximate the whole, or part of, the original method. An important part of the approach is to show how improvements can be introduced. In line with the presented philosophy, this is done by extending the traditional method in several ways followed by ANN approximation of the modified algorithms. The extensions are based on a windowed version of the nearest neighbour algorithm. We show that the improvements carry over to the ANN implementations. We further investigate the stability of the solutions by modifying the training set. It is shown that the errors do not change significantly. This also holds true for the ANN approximations providing confidence that the two‐step strategy is robust. [ABSTRACT FROM AUTHOR]

Wireless mesh networks are a special class of wireless networks that are implemented as a collection of radio nodes in a mesh pattern or topology. Unlike MANETs, the mobility of nodes is very less in the topology. Quality of service is an essential metric in the performance of mesh networks which are attributed to several parameters including optimal routing through shortest path, ability for other nodes to communicate even if a particular node in the mesh fails, minimization of packet loss and time delay, computational complexity and cost, energy. This research paper is focused toward minimization of energy taken as the objective function and a five-stage neural network is used and trained after optimizing with a genetic algorithm. The experiments have been conducted in NS2 and Qualnet environment with a varying number of mesh routers and energy computed. The performance of energy savings has been compared against conventional routing techniques like AODV and a bee colony optimization technique presented in the literature. An energy savings of 51% have been reported in the paper justifying the superiority of the hybrid G-ANN algorithm. [ABSTRACT FROM AUTHOR]

Analyzing and predicting the trend of price fluctuation has been receiving more and more attention, as price risk has become the focus of risk control research in heating oil futures market. A novel time series prediction model combined with the complex network method is put forward in the paper. First of all, this paper counts the cumulative time interval of different nodes in the network, and fits its growth trend with the Fourier model. Then a novel price fluctuation prediction model is established based on the effective information such as some topology properties extracted from the network. The results show that the Fourier model can predict the emergence time of new nodes in the next stage, and the established price fluctuation prediction model can infer the names of nodes in the prediction interval, so as to determine the forward-looking behavior of price evolution. Besides, liken to the NAR neural network, the prediction results obtained by the proposed method also show superiority, which has important theoretical value and academic significance for early warning and prediction of price behavior in the heating oil futures market. [ABSTRACT FROM AUTHOR]

The 3D measurement system based on line-structured light uses a camera to capture laser stripes due to changing in the shape of an object, and uses the acquired pixel coordinates for 3D reconstruction. System calibration is an important step in 3D measurement. The current camera calibration algorithm research mainly focuses on improving the algorithm itself, and there is less research on the influence of external factors. This paper proposes a coplanar hybrid calibration algorithm based on the error screening model by combining the error screening model, mathematical model and neural network model. It is mainly divided into two steps. The first step is to use the radial array constraint calibration algorithm based on the error screening model to solve the camera's internal and external parameters. The second step uses the camera internal and external parameters obtained in the first step to convert the pixel coordinates into real three-dimensional coordinates, and compares the calculated three-dimensional coordinates with the actual coordinates. Using machine learning to establish a compensation network, get a compensation function, and use the resulting 3D world coordinates to perform point cloud stitching. Experiments show that compared with the traditional calibration algorithm, the calibration algorithm has a small error and reduces the calibration error by about 6.5%. [ABSTRACT FROM AUTHOR]

Neural network learning is usually time-consuming since backpropagation needs to compute full gradients and backpropagate them across multiple layers. Despite its success of existing works in accelerating propagation through sparseness, the relevant theoretical characteristics remain under-researched and empirical studies found that they suffer from the loss of information contained in unpropagated gradients. To tackle these problems, this paper presents a unified sparse backpropagation framework and provides a detailed analysis of its theoretical characteristics. Analysis reveals that when applied to a multilayer perceptron, our framework essentially performs gradient descent using an estimated gradient similar enough to the true gradient, resulting in convergence in probability under certain conditions. Furthermore, a simple yet effective algorithm named m emorized s parse b ack p ropagation (MSBP) is proposed to remedy the problem of information loss by storing unpropagated gradients in memory for learning in the next steps. Experimental results demonstrate that the proposed MSBP is effective to alleviate the information loss in traditional sparse backpropagation while achieving comparable acceleration. [ABSTRACT FROM AUTHOR]

In this paper, we propose a hybrid method to forecast the stock prices called High-order-fuzzy-fluctuation-Trends-based Back Propagation(HTBP)Neural Network model. First, we compare each value of the historical training data with the previous day's value to obtain a fluctuation trend time series (FTTS). On this basis, the FTTS blur into fuzzy time series (FFTS) based on the fluctuation of the increasing, equality, decreasing amplitude and direction. Since the relationship between FFTS and future wave trends is nonlinear, the HTBP neural network algorithm is used to find the mapping rules in the form of self-learning. Finally, the results of the algorithm output are used to predict future fluctuations. The proposed model provides some innovative features:(1)It combines fuzzy set theory and neural network algorithm to avoid overfitting problems existed in traditional models. (2)BP neural network algorithm can intelligently explore the internal rules of the actual existence of sequential data, without the need to analyze the influence factors of specific rules and the path of action. (3)The hybrid modal can reasonably remove noises from the internal rules by proper fuzzy treatment. This paper takes the TAIEX data set of Taiwan stock exchange as an example, and compares and analyzes the prediction performance of the model. The experimental results show that this method can predict the stock market in a very simple way. At the same time, we use this method to predict the Shanghai stock exchange composite index, and further verify the effectiveness and universality of the method. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach to the localization of moving targets in a complex environment based on the measurement of the perturbations induced by the target presence on an independently‐generated time‐varying electromagnetic field. Field perturbations are measured via a set of sensors deployed over the domain of interest and used to detect and track a possible target by resorting to a particle Bernoulli filter (PBF). To comply with real‐time operation, the PBF works along with an artificial neural network (ANN) model of the environment trained offline via finite elements (FEs). The performance of the proposed algorithm is assessed via simulation experiments. [ABSTRACT FROM AUTHOR]

Neural network algorithms are gradually applied to the field of chemical informatics. In this paper, the neural network models are used to predict the properties of compounds based on Molecular Information. We predict the aqueous solubility of compounds, and evaluate the prediction results of the Neural Networks including CNN, RNN, DNN, SNN. The performance of the models in predicting the solubility is able to meet or exceed the predicted effect of the method based on the molecular structure (ESOL). DNN model performance is of more accuracy, and RNN performance is of better stability. This method can directly avoid complex molecular structure characterisation, and provide a convenient and flexible way to predict properties of compounds. [ABSTRACT FROM AUTHOR]

Leveraging smart metering solutions to support energy efficiency on the individual household level poses novel research challenges in monitoring usage and providing accurate load forecasting. Forecasting electricity usage is an especially important component that can provide intelligence to smart meters. In this paper, we propose an enhanced approach for load forecasting at the household level. The impacts of residents’ daily activities and appliance usages on the power consumption of the entire household are incorporated to improve the accuracy of the forecasting model. The contributions of this paper are threefold: (1) we addressed short-term electricity load forecasting for 24 hours ahead, not on the aggregate but on the individual household level, which fits into the Residential Power Load Forecasting (RPLF) methods; (2) for the forecasting, we utilized a household specific dataset of behaviors that influence power consumption, which was derived using segmentation and sequence mining algorithms; and (3) an extensive load forecasting study using different forecasting algorithms enhanced by the household activity patterns was undertaken. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we investigate the supervised backpropagation training of multilayer neural networks from a dynamical systems point of view. We discuss some links with the qualitative theory of differential equations and introduce the overfly algorithm to tackle the local minima problem. Our approach is based on the existence of first integrals of the generalised gradient system with build-in dissipation. [ABSTRACT FROM AUTHOR]

Electroencephalogram (EEG) is a common base signal used to monitor brain activities and diagnose sleep disorders. Manual sleep stage scoring is a time-consuming task for sleep experts and is limited by inter-rater reliability. In this paper, we propose an automatic sleep stage annotation method called SleepEEGNet using a single-channel EEG signal. The SleepEEGNet is composed of deep convolutional neural networks (CNNs) to extract time-invariant features, frequency information, and a sequence to sequence model to capture the complex and long short-term context dependencies between sleep epochs and scores. In addition, to reduce the effect of the class imbalance problem presented in the available sleep datasets, we applied novel loss functions to have an equal misclassified error for each sleep stage while training the network. We evaluated the performance of the proposed method on different single-EEG channels (i.e., Fpz-Cz and Pz-Oz EEG channels) from the Physionet Sleep-EDF datasets published in 2013 and 2018. The evaluation results demonstrate that the proposed method achieved the best annotation performance compared to current literature, with an overall accuracy of 84.26%, a macro F1-score of 79.66% and κ = 0.79. Our developed model can be applied to other sleep EEG signals and aid the sleep specialists to arrive at an accurate diagnosis. The source code is available at . [ABSTRACT FROM AUTHOR]

This paper presents two algorithms, both based on the time of flight (ToF) of scattered waves, to locate defect in an anisotropic woven-fabric carbon fiber reinforced polymer (CFRP) plate. The first algorithm uses a probabilistic approach by constructing a probability matrix. Each element of this matrix is associated with a location and represents the probability of existing a defect at the corresponding spatial coordinates of the plate. For the probability matrix method, localization results are influenced by manually chosen parameters and by the anisotropy of the CFRP plate. The second algorithm, based on artificial neural networks (ANNs), enables us to improve the accuracy of locating defects. With this ANN method, the anisotropic feature can be “learned” by the neural network from training data. The ToF of scattered waves obtained from three sensor pairs were used directly as inputs of the neural network. The spatial coordinates of the defect are the ANN outputs. The difficulty of obtaining sufficient experimental data for ANN training was surpassed by using added blocks on the surface of the plate under test to simulate defects. This scheme was validated and proved to be effective by comparing the scattered waves from a delamination and from an added block. Conclusions have been drawn by comparing the two methods. The localization results obtained by the ANN algorithm are proved to be better. [ABSTRACT FROM AUTHOR]

Antenna selection in Multiple-Input Multiple-Output (MIMO) systems has attracted increasing attention due to the challenge of keeping a balance between communication performance and computational complexity. Recently, deep learning based methods have achieved promising performance in many application fields. This paper proposed a deep learning (DL) based antenna selection technique. First, we generated the label of training antenna systems by maximizing the channel capacity. Then, we adopted the deep convolutional neural network (CNN) on the channel matrices to explicitly exploit the massive latent cues of attenuation coefficients. Finally, we used the adopted CNN to assign the class label and then select the optimal antenna subset. Experimental results demonstrate that our method can achieve better performance than the state-of-the-art baselines for data-driven based antenna selection. [ABSTRACT FROM AUTHOR]

In general, the stock prices of the same industry have a similar trend, but those of different industries do not. When investing in stocks of different industries, one should select the optimal model from lots of trading models for each industry because any model may not be suitable for capturing the stock trends of all industries. However, the study has not been carried out at present. In this paper, firstly we select 424 S&P 500 index component stocks (SPICS) and 185 CSI 300 index component stocks (CSICS) as the research objects from 2010 to 2017, divide them into 9 industries such as finance and energy respectively. Secondly, we apply 12 widely used machine learning algorithms to generate stock trading signals in different industries and execute the back-testing based on the trading signals. Thirdly, we use a non-parametric statistical test to evaluate whether there are significant differences among the trading performance evaluation indicators (PEI) of different models in the same industry. Finally, we propose a series of rules to select the optimal models for stock investment of every industry. The analytical results on SPICS and CSICS show that we can find the optimal trading models for each industry based on the statistical tests and the rules. Most importantly, the PEI of the best algorithms can be significantly better than that of the benchmark index and “Buy and Hold” strategy. Therefore, the algorithms can be used for making profits from industry stock trading. [ABSTRACT FROM AUTHOR]

Highlights • Method simulates effects of launch conditions on rail-gun current and armature speed. • The input five parameters are closely related to the rail-gun launchment. • Three neural networks predict the rail-gun current and armature speed with low error. • General Regression neural network is optimal on error, time cost, parameter design. Abstract In this paper, based on the advantages of artificial neural network, such as good tolerance of data noise, strong ability of nonlinear mapping, multi-dimensional input variables, fast operation, low error, etc., a method of using artificial neural network for data prediction is proposed for the research of rail-gun. The results show that it is feasible to use the Back Propagation Neural Network, the Radial Basis Function Neural Network and the General Regression Neural Network to realize the method of prediction and simulation of the rail-gun current and the armature speed curve through relevant parameters. The General Regression Neural Network has superiority in error performance and time cost of neural network training and simulation. [ABSTRACT FROM AUTHOR]

The formation of manufacturing cells forms the backbone of designing a cellular manufacturing system. In this paper, we present a novel intelligent particle swarm optimization algorithm for the cell formation problem. The proposed solution method benefits from the advantages of particle swarm optimization algorithm (PSO) and self-organization map neural networks by combining artificial individual intelligence and swarm intelligence. Numerical examples demonstrate that the proposed intelligent particle swarm optimization algorithm significantly outperforms PSO and yields better solutions than the best solutions existed in the literature of cell formation. The application of the proposed approach is examined in a case problem where real data is utilized for cell reconfiguration of an actual company involved in agricultural manufacturing sector. [ABSTRACT FROM AUTHOR]

Homology Continuity is a fundamental property of the nature, but few of the traditional pattern recognition algorithms were aware of it. Firstly, this paper gives a brief description to the Principle of Homology Continuity (PHC), and tries to mathematically redefine it. Then, we introduce a PHC-based pattern learning method — Geometrical Covering Learning (GCL), following the Hyper sausage neural network as an instance of GCL. Lastly, we propose a GCL solution to the “two-spirals” pattern recognition problem. The final experimental results show that the new method is feasible and efficient. [ABSTRACT FROM AUTHOR]

In palynology, the visual classification of pollen grains from different species is a hard task which is usually tackled by human operators using microscopes. Its complete automatization would save a high quantity of resources and provide valuable improvements especially for allergy-related information systems, but also for other application fields as paleoclimate reconstruction, quality control of honey based products, collection of evidences in criminal investigations or fabric dating and tracking. This paper presents three state-of-the-art deep learning classification methods applied to the recently published POLEN23E image dataset. The three methods make use of convolutional neural networks: the first one is strictly based on the idea of transfer learning, the second one is based on feature extraction and the third one represents a hybrid approach, combining transfer learning and feature extraction. The results from the three methods are indeed very good, reaching over 97% correct classification rates in images not previously seen by the models, where other authors reported around 70. [ABSTRACT FROM AUTHOR]

Semantic segmentation tasks can be well modeled by Markov Random Field (MRF). This paper addresses semantic segmentation by incorporating high-order relations and mixture of label contexts into MRF. Unlike previous works that optimized MRFs using iterative algorithm, we solve MRF by proposing a Convolutional Neural Network (CNN), namely Deep Parsing Network (DPN), which enables deterministic end-to-end computation in a single forward pass. Specifically, DPN extends a contemporary CNN to model unary terms and additional layers are devised to approximate the mean field (MF) algorithm for pairwise terms. It has several appealing properties. First, different from the recent works that required many iterations of MF during back-propagation, DPN is able to achieve high performance by approximating one iteration of MF. Second, DPN represents various types of pairwise terms, making many existing models as its special cases. Furthermore, pairwise terms in DPN provide a unified framework to encode rich contextual information in high-dimensional data, such as images and videos. Third, DPN makes MF easier to be parallelized and speeded up, thus enabling efficient inference. DPN is thoroughly evaluated on standard semantic image/video segmentation benchmarks, where a single DPN model yields state-of-the-art segmentation accuracies on PASCAL VOC 2012, Cityscapes dataset and CamVid dataset. [ABSTRACT FROM AUTHOR]

An accurate battery remaining useful life (RUL) estimation can facilitate the design of a reliable battery system as well as the safety and reliability of actual operation. A reasonable definition and an effective prediction algorithm are indispensable for the achievement of an accurate RUL estimation result. In this paper, the analysis of battery terminal voltage curves under different cycle numbers during charge process is utilized for RUL definition. Moreover, the relationship between RUL and charge curve is simulated by feed forward neural network (FFNN) for its simplicity and effectiveness. Considering the nonlinearity of lithium-ion charge curve, importance sampling (IS) is employed for FFNN input selection. Based on these results, an online approach using FFNN and IS is presented to estimate lithium-ion battery RUL in this paper. Experiments and numerical comparisons are conducted to validate the proposed method. The results show that the FFNN with IS is an accurate estimation method for actual operation. [ABSTRACT FROM AUTHOR]

During past 50 years, the markets have been the object of study. Technical and fundamental indicators have been used to try to predict the behavior of the market and then execute buying or selling orders. Neural networks are currently being used with good results although they can be useless after a period of time. This paper proposes an algorithm that combines bioinspired techniques to maximize the hits in the prediction rates. The proposal shown in this paper relies in an ANN to achieve these goals. The differential factors of this approach are the election of the ANN structure with grammatical swarm and the training process through the use of HydroPSO. Also a grammatical swarm algorithm is used to generate trading rules as this method shows better results than the first approach. This combination of techniques provides an automatic way to define the most suitable bioinspired model for the instrument in our analysis. [ABSTRACT FROM AUTHOR]

The spiking neural networks (SNNs) are the third generation of neural networks and perform remarkably well in cognitive tasks such as pattern recognition. The spike emitting and information processing mechanisms found in biological cognitive systems motivate the application of the hierarchical structure and temporal encoding mechanism in spiking neural networks, which have exhibited strong computational capability. However, the hierarchical structure and temporal encoding approach require neurons to process information serially in space and time respectively, which reduce the training efficiency significantly. For training the hierarchical SNNs, most existing methods are based on the traditional back-propagation algorithm, inheriting its drawbacks of the gradient diffusion and the sensitivity on parameters. To keep the powerful computation capability of the hierarchical structure and temporal encoding mechanism, but to overcome the low efficiency of the existing algorithms, a new training algorithm, the Normalized Spiking Error Back Propagation (NSEBP) is proposed in this paper. In the feedforward calculation, the output spike times are calculated by solving the quadratic function in the spike response model instead of detecting postsynaptic voltage states at all time points in traditional algorithms. Besides, in the feedback weight modification, the computational error is propagated to previous layers by the presynaptic spike jitter instead of the gradient decent rule, which realizes the layer-wised training. Furthermore, our algorithm investigates the mathematical relation between the weight variation and voltage error change, which makes the normalization in the weight modification applicable. Adopting these strategies, our algorithm outperforms the traditional SNN multi-layer algorithms in terms of learning efficiency and parameter sensitivity, that are also demonstrated by the comprehensive experimental results in this paper. [ABSTRACT FROM AUTHOR]

In this paper, neural algorithms, including the multi-layered perceptron (MLP) differential approximator, generalized hybrid power series, discrete Hopfield neural network, and the hybrid numerical, are used for constructing models that incorporate a priori knowledge in the form of differential equations for dynamic engineering processes. The properties of these approaches are discussed and compared to each other in terms of efficiency and accuracy. The presented algorithms have a number of advantages over other traditional mesh-based methods such as reduction of the computational cost, speed up of the execution time, and data integration with the a priori knowledge. Furthermore, the presented techniques are applicable when the differential equations governing a system or dynamic engineering process are not fully understood. The proposed algorithms learn to compute the unknown or free parameters of the equation from observations of the process behavior, hence a more precise theoretical description of the process is obtained. Additionally, there will be no need to solve the differential equation each time the free parameters change. The parallel nature of the approaches outlined in this paper make them attractive for parallel implementation in dynamic engineering processes. [ABSTRACT FROM AUTHOR]

Deep learning techniques, e.g., Convolutional Neural Networks (CNNs), have been explosively applied to the research in the fields of information retrieval and natural language processing. However, few research efforts have addressed semantic indexing with deep learning. The use of semantic indexing in the biomedical literature has been limited for several reasons. For instance, MEDLINE citations contain a large number of semantic labels from automatically annotated MeSH terms, and for a great deal of the literature, only the information of the title and the abstract is readily available. In this paper, we propose a Boltzmann Convolutional neural network framework (B-CNN) for biomedicine semantic indexing. In our hybrid learning framework, the CNN can adaptively deal with features of documents that have sequence relationships, and can capture context information accordingly; the Deep Boltzmann Machine (DBM) merges global (the entity in each document) and local information through its training with undirected connections. Additionally, we have designed a hierarchical coarse to fine style indexing structure for learning and classifying documents, and a novel feature extension approach with word sequence embedding and Wikipedia categorization. Comparative experiments were conducted for semantic indexing of biomedical abstract documents; these experiments verified the encouraged performance of our B-CNN model. [ABSTRACT FROM AUTHOR]

This paper presents a new method to reduce the computational cost when using Neural Networks as Language Models, during recognition, in some particular scenarios. It is based on a Neural Network that considers input contexts of different length in order to ease the use of a fallback mechanism together with the precomputation of softmax normalization constants for these inputs. The proposed approach is empirically validated, showing their capability to emulate lower order N-grams with a single Neural Network. A machine translation task shows that the proposed model constitutes a good solution to the normalization cost of the output softmax layer of Neural Networks, for some practical cases, without a significant impact in performance while improving the system speed. [ABSTRACT FROM AUTHOR]

A residual networks family with hundreds or even thousands of layers dominates major image recognition tasks, but building a network by simply stacking residual blocks inevitably limits its optimization ability. This paper proposes a novel residual network architecture, residual networks of residual networks (RoR), to dig the optimization ability of residual networks. RoR substitutes optimizing residual mapping of residual mapping for optimizing original residual mapping. In particular, RoR adds levelwise shortcut connections upon original residual networks to promote the learning capability of residual networks. More importantly, RoR can be applied to various kinds of residual networks (ResNets, Pre-ResNets, and WRN) and significantly boost their performance. Our experiments demonstrate the effectiveness and versatility of RoR, where it achieves the best performance in all residual-network-like structures. Our RoR-3-WRN58-4 + SD models achieve new state-of-the-art results on CIFAR-10, CIFAR-100, and SVHN, with the test errors of 3.77%, 19.73%, and 1.59%, respectively. RoR-3 models also achieve state-of-the-art results compared with ResNets on the ImageNet data set. [ABSTRACT FROM AUTHOR]

Combining a deep neural network with fuzzy theory, this paper proposes an advertising click-through rate (CTR) prediction approach based on a fuzzy deep neural network (FDNN). In this approach, fuzzy Gaussian-Bernoulli restricted Boltzmann machine (FGBRBM) is first applied to input raw data from advertising datasets. Next, fuzzy restricted Boltzmann machine (FRBM) is used to construct the fuzzy deep belief network (FDBN) with the unsupervised method layer by layer. Finally, fuzzy logistic regression (FLR) is utilized for modeling the CTR. The experimental results show that the proposed FDNN model outperforms several baseline models in terms of both data representation capability and robustness in advertising click log datasets with noise. [ABSTRACT FROM AUTHOR]

In this paper, we seek to improve deep neural networks by generalizing the pooling operations that play a central role in the current architectures. We pursue a careful exploration of approaches to allow pooling to learn and to adapt to complex and variable patterns. The two primary directions lie in: (1) learning a pooling function via (two strategies of) combining of max and average pooling, and (2) learning a pooling function in the form of a tree-structured fusion of pooling filters that are themselves learned. In our experiments every generalized pooling operation we explore improves performance when used in place of average or max pooling. We experimentally demonstrate that the proposed pooling operations provide a boost in invariance properties relative to conventional pooling and set the state of the art on several widely adopted benchmark datasets. These benefits come with only a light increase in computational overhead during training (ranging from additional 5 to 15 percent in time complexity) and a very modest increase in the number of model parameters (e.g., additional 1, 9, and 27 parameters for mixed, gated, and 2-level tree pooling operators, respectively). To gain more insights about our proposed pooling methods, we also visualize the learned pooling masks and the embeddings of the internal feature responses for different pooling operations. Our proposed pooling operations are easy to implement and can be applied within various deep neural network architectures. [ABSTRACT FROM PUBLISHER]

Recent research on speech segregation and music fingerprinting has led to improvements in speech segregation and music identification algorithms. Speech and music segregation generally involves the identification of music followed by speech segregation. However, music segregation becomes a challenging task in the presence of noise. This paper proposes a novel method of speech segregation for unlabelled stationary noisy audio signals using the deep belief network (DBN) model. The proposed method successfully segregates a music signal from noisy audio streams. A recurrent neural network (RNN)-based hidden layer segregation model is applied to remove stationary noise. Dictionary-based fisher algorithms are employed for speech classification. The proposed method is tested on three datasets (TIMIT, MIR-1K, and MusicBrainz), and the results indicate the robustness of proposed method for speech segregation. The qualitative and quantitative analysis carried out on three datasets demonstrate the efficiency of the proposed method compared to the state-of-the-art speech segregation and classification-based methods. [ABSTRACT FROM AUTHOR]

Advanced Metering Infrastructure (AMI) realizes a two-way communication of electricity data through by interconnecting with a computer network as the core component of the smart grid. Meanwhile, it brings many new security threats and the traditional intrusion detection method can’t satisfy the security requirements of AMI. In this paper, an intrusion detection system based on Online Sequence Extreme Learning Machine (OS-ELM) is established, which is used to detecting the attack in AMI and carrying out the comparative analysis with other algorithms. Simulation results show that, compared with other intrusion detection methods, intrusion detection method based on OS-ELM is more superior in detection speed and accuracy. [ABSTRACT FROM AUTHOR]

This paper presents a neural network-based method to generate multiple images with different exposures from a single input low dynamic range (LDR) image for improved high dynamic range (HDR) imaging. The proposed algorithm consists of three steps: 1) 2-D histogram estimation; 2) neural network-based LDR images estimation; and 3) generation of an optimal set of differently exposed images. The proposed method first generates image features by estimating a patched-based 2-D histogram. The extracted features are used in an input layer of the neural network, which plays a role to select an optimal set of LDR images. A set of LDR images is generated using a curvature-based contrast enhancement method. Experimental results show that the proposed method can generate an optimal set of LDR images using neural network and provide improved HDR images. In addition, the proposed method can be implemented as a preprocessing step in most existing HDR frameworks. The proposed HDR approach is considered as a single-input method that gives almost the same performance to multiple image-based HDR method. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a new automatic hyperparameter selection approach for determining the optimal network configuration (network structure and hyperparameters) for deep neural networks using particle swarm optimization (PSO) in combination with a steepest gradient descent algorithm. In the proposed approach, network configurations were coded as a set of real-number m-dimensional vectors as the individuals of the PSO algorithm in the search procedure. During the search procedure, the PSO algorithm is employed to search for optimal network configurations via the particles moving in a finite search space, and the steepest gradient descent algorithm is used to train the DNN classifier with a few training epochs (to find a local optimal solution) during the population evaluation of PSO. After the optimization scheme, the steepest gradient descent algorithm is performed with more epochs and the final solutions (pbest and gbest) of the PSO algorithm to train a final ensemble model and individual DNN classifiers, respectively. The local search ability of the steepest gradient descent algorithm and the global search capabilities of the PSO algorithm are exploited to determine an optimal solution that is close to the global optimum. We constructed several experiments on hand-written characters and biological activity prediction datasets to show that the DNN classifiers trained by the network configurations expressed by the final solutions of the PSO algorithm, employed to construct an ensemble model and individual classifier, outperform the random approach in terms of the generalization performance. Therefore, the proposed approach can be regarded an alternative tool for automatic network structure and parameter selection for deep neural networks. [ABSTRACT FROM AUTHOR]

Following the great success of deep convolutional neural networks (CNNs) in computer vision, this paper proposes a complex-valued CNN (CV-CNN) specifically for synthetic aperture radar (SAR) image interpretation. It utilizes both amplitude and phase information of complex SAR imagery. All elements of CNN including input-output layer, convolution layer, activation function, and pooling layer are extended to the complex domain. Moreover, a complex backpropagation algorithm based on stochastic gradient descent is derived for CV-CNN training. The proposed CV-CNN is then tested on the typical polarimetric SAR image classification task which classifies each pixel into known terrain types via supervised training. Experiments with the benchmark data sets of Flevoland and Oberpfaffenhofen show that the classification error can be further reduced if employing CV-CNN instead of conventional real-valued CNN with the same degrees of freedom. The performance of CV-CNN is comparable to that of existing state-of-the-art methods in terms of overall classification accuracy. [ABSTRACT FROM PUBLISHER]

In a modern small gasoline engine fuel injection system, the load of the engine is estimated based on the measurement of the manifold absolute pressure (MAP) sensor, which took place in the intake manifold. This paper present a more economical approach on estimating the MAP by using only the measurements of the throttle position and engine speed, resulting in lower implementation cost. The estimation was done via two-stage multilayer feed-forward neural network by combining Levenberg-Marquardt (LM) algorithm, Bayesian Regularization (BR) algorithm and Particle Swarm Optimization (PSO) algorithm. Based on the results found in 20 runs, the second variant of the hybrid algorithm yields a better network performance than the first variant of hybrid algorithm, LM, LM with BR and PSO by estimating the MAP closely to the simulated MAP values. By using a valid experimental training data, the estimator network that trained with the second variant of the hybrid algorithm showed the best performance among other algorithms when used in an actual retrofit fuel injection system (RFIS). The performance of the estimator was also validated in steady-state and transient condition by showing a closer MAP estimation to the actual value. [ABSTRACT FROM AUTHOR]

In this article, we propose a new data mining algorithm, by which one can both capture the non-linearity in data and also find the best subset model. To produce an enhanced subset of the original variables, a preferred selection method should have the potential of adding a supplementary level of regression analysis that would capture complex relationships in the data via mathematical transformation of the predictors and exploration of synergistic effects of combined variables. The method that we present here has the potential to produce an optimal subset of variables, rendering the overall process of model selection more efficient. This algorithm introduces interpretable parameters by transforming the original inputs and also a faithful fit to the data. The core objective of this paper is to introduce a new estimation technique for the classical least square regression framework. This new automatic variable transformation and model selection method could offer an optimal and stable model that minimizes the mean square error and variability, while combining all possible subset selection methodology with the inclusion variable transformations and interactions. Moreover, this method controls multicollinearity, leading to an optimal set of explanatory variables. [ABSTRACT FROM AUTHOR]

In this paper, we propose an automated detection algorithm for the localization of apexes and points on the prongs of hyperbolic reflection incurred as a result of GPR scanning technology. The objects of interest encompass cylindrical underground utilities that have a distinctive form of hyperbolic reflection in radargram. Algorithm involves application of trained neural network to analyze radargram in the form of raster image, resulting with extracted segments of interest that contain hyperbolic reflections. This significantly reduces the amount of data for further analysis. Extracted segments represent the zone for localization of apices. This is followed by extraction of points on prongs of hyperbolic reflections which is carried out until stopping criterion is satisfied, regardless the borders of segment of interest. In final step a classification of false hyperbolic reflections caused by the constructive interference and their removal is done. The algorithm is implemented in MATLAB environment. There are several advantages of the proposed algorithm. It can successfully recognize true hyperbolic reflections in radargram images and extracts coordinates, with very low rate of false detections and without prior knowledge about the number of hyperbolic reflections or buried utilities. It can be applied to radargrams containing single and multiple hyperbolic reflections, intersected, distorted, as well as incomplete (asymmetric) hyperbolic reflections, all in the presence of higher level of noise. Special feature of algorithm is developed procedure for analysis and removal of false hyperbolic reflections generated by the constructive interference from reflectors associated with the utilities. Algorithm was tested on a number of synthetic and radargram acquired in the field survey. To illustrate the performances of the proposed algorithm, we present the characteristics of the algorithm through five representative radargrams obtained in real conditions. In these examples we present different acquisition scenarios by varying the number of buried objects, their disposition, size, and level of noise. Example with highest complexity was tested also as a synthetic radargram generated by gprMax. Processing time in examples with one or two hyperbolic reflections is from 0.1 to 0.3 s, while for the most complex examples it is from 2.2 to 4.9 s. In general, the obtained experimental results show that the proposed algorithm exhibits promising performances both in terms of utility detection and processing speed of the algorithm. [ABSTRACT FROM AUTHOR]

In this paper, we present a robust text detection approach in natural images which is based on region proposal mechanism. A powerful low-level detector named saliency enhanced-MSER extended from the widely-used MSER is proposed by incorporating saliency detection methods, which ensures a high recall rate. Given a natural image, character candidates are extracted from three channels in a perception-based illumination invariant color space by saliency-enhanced MSER algorithm. A discriminative convolutional neural network (CNN) is jointly trained with multi-level information including pixel-level and character-level information as character candidate classifier. Each image patch is classified as strong text, weak text and non-text by double threshold filtering instead of conventional one-step classification, leveraging confident scores obtained via CNN. To further prune non-text regions, we develop a recursive neighborhood search algorithm to track credible texts from weak text set. Finally, characters are grouped into text lines using heuristic features such as spatial location, size, color, and stroke width. We compare our approach with several state-of-the-art methods, and experiments show that our method achieves competitive performance on public datasets ICDAR 2011 and ICDAR 2013. [ABSTRACT FROM AUTHOR]

Modern computing architecture based on the separation of memory and processing leads to a well known problem called the von Neumann bottleneck, a restrictive limit on the data bandwidth between CPU and RAM. This paper introduces a new approach to computing we call AHaH computing where memory and processing are combined. The idea is based on the attractor dynamics of volatile dissipative electronics inspired by biological systems, presenting an attractive alternative architecture that is able to adapt, self-repair, and learn from interactions with the environment. We envision that both von Neumann and AHaH computing architectures will operate together on the same machine, but that the AHaH computing processor may reduce the power consumption and processing time for certain adaptive learning tasks by orders of magnitude. The paper begins by drawing a connection between the properties of volatility, thermodynamics, and Anti-Hebbian and Hebbian (AHaH) plasticity. We show how AHaH synaptic plasticity leads to attractor states that extract the independent components of applied data streams and how they form a computationally complete set of logic functions. After introducing a general memristive device model based on collections of metastable switches, we show how adaptive synaptic weights can be formed from differential pairs of incremental memristors. We also disclose how arrays of synaptic weights can be used to build a neural node circuit operating AHaH plasticity. By configuring the attractor states of the AHaH node in different ways, high level machine learning functions are demonstrated. This includes unsupervised clustering, supervised and unsupervised classification, complex signal prediction, unsupervised robotic actuation and combinatorial optimization of procedures–all key capabilities of biological nervous systems and modern machine learning algorithms with real world application. [ABSTRACT FROM AUTHOR]

The present paper covers the issue of training of an artificial neural network in an intelligent diagnostic system whose purpose is to evaluate repairable technical objects. The structure of the diagnostic system was characterized, and the measurement and diagnostic subsystems were described. An artificial neural network is an important element in an intelligent diagnostic subsystem. The structure, the algorithm, the organization of a neural network and the basic relations that describe its work were presented. The information presented in the form of the vectors of diagnostic signals, and their standard vectors constitute the primary information base used in 'DIAG' computer program. Training of an artificial neural network is an important aspect that is presented in the paper. The issue concerning these problems is not presented in the literature. Training of a network was presented on the grounds of teaching vectors, which are determined in a diagnostic system in the process of a simulation of a specific state in the object examined. An example of training of a network was presented in a diagnostic system which evaluates a control system of the operation of a car engine. Appropriate connections were presented for the purpose of a qualitative assessment of the training process of a neural network. [ABSTRACT FROM AUTHOR]

In this paper, a new program for protein secondary structure prediction is proposed. The program, which is called NN-Pred, allows multiple sequences to be submitted and it returns predictions from five secondary structure prediction algorithms. In addition, NN-Pred calculates a consensus prediction, which is based on a neural network strategy that is used in this paper to improve the prediction accuracy. NN-Pred was obtained by using a methodology called consensus strategy, which tries to make a better prediction by integrating some of the most remarkable existing techniques. The NN-Pred program provides a three-state (alpha-helix, beta-sheet, and other) prediction of secondary structure. According to the test sets, the prediction accuracy of NN-Pred is at least 70%, surpassing most of the existing methods. The experiments showed that neural networks can be used as a consensus strategy to producing accurate models for protein secondary prediction. [ABSTRACT FROM AUTHOR]

Gramophone records were the main recording medium for more than seven decades and regained widespread popularity over the past several years. Being an analog storage medium, gramophone records are subject to distortions caused by scratches, dust particles, degradation, and other means of improper handling. The observed noise often leads to an unpleasant listening experience and requires a filtering process to remove the unwanted disruptions and improve the audio quality. This paper proposes a novel approach that employs various feed forward time delay artificial neural networks to detect and reconstruct noise in musical sound waves. A set of 800 songs from eight different genres were used to validate the performance of the neural networks. The performance was analyzed according to the outlier detection and interpolation accuracy, the computational time and the tradeoff between the accuracy and the time. The empirical results of both detection and reconstruction neural networks were compared to a number of other algorithms, including various statistical measurements, duplication approaches, trigonometric processes, polynomials, and time series models. It was found that the neural networks’ outlier detection accuracy was slightly lower than some of the other noise identification algorithms, but achieved a more efficient tradeoff by detecting most of the noise in real time. The reconstruction process favored neural networks with an increase in the interpolation accuracy compared to other widely used time series models. It was also found that certain genres such as classical, country, and jazz music were interpolated more accurately. Volatile signals, such as electronic, metal, and pop music were more challenging to reconstruct and were substantially better interpolated using neural networks than the other examined algorithms. [ABSTRACT FROM PUBLISHER]

In this paper, we focus on tackling the problem of automatic accurate localization of detected objects in high-resolution remote sensing images. The two major problems for object localization in remote sensing images caused by the complex context information such images contain are achieving generalizability of the features used to describe objects and achieving accurate object locations. To address these challenges, we propose a new object localization framework, which can be divided into three processes: region proposal, classification, and accurate object localization process. First, a region proposal method is used to generate candidate regions with the aim of detecting all objects of interest within these images. Then, generic image features from a local image corresponding to each region proposal are extracted by a combination model of 2-D reduction convolutional neural networks (CNNs). Finally, to improve the location accuracy, we propose an unsupervised score-based bounding box regression (USB-BBR) algorithm, combined with a nonmaximum suppression algorithm to optimize the bounding boxes of regions that detected as objects. Experiments show that the dimension-reduction model performs better than the retrained and fine-tuned models and the detection precision of the combined CNN model is much higher than that of any single model. Also our proposed USB-BBR algorithm can more accurately locate objects within an image. Compared with traditional features extraction methods, such as elliptic Fourier transform-based histogram of oriented gradients and local binary pattern histogram Fourier, our proposed localization framework shows robustness when dealing with different complex backgrounds. [ABSTRACT FROM PUBLISHER]

Style transfer is a process of migrating a style from a given image to the content of another, synthesizing a new image, which is an artistic mixture of the two. Recent work on this problem adopting convolutional neural-networks (CNN) ignited a renewed interest in this field, due to the very impressive results obtained. There exists an alternative path toward handling the style transfer task, via the generalization of texture synthesis algorithms. This approach has been proposed over the years, but its results are typically less impressive compared with the CNN ones. In this paper, we propose a novel style transfer algorithm that extends the texture synthesis work of Kwatra et al. (2005), while aiming to get stylized images that are closer in quality to the CNN ones. We modify Kwatra’s algorithm in several key ways in order to achieve the desired transfer, with emphasis on a consistent way for keeping the content intact in selected regions, while producing hallucinated and rich style in others. The results obtained are visually pleasing and diverse, shown to be competitive with the recent CNN style transfer algorithms. The proposed algorithm is fast and flexible, being able to process any pair of content + style images. [ABSTRACT FROM PUBLISHER]

Music emotion recognition (MER) field rapidly expanded in the last decade. Many new methods and new audio features are developed to improve the performance of MER algorithms. However, it is very difficult to compare the performance of the new methods because of the data representation diversity and scarcity of publicly available data. In this paper, we address these problems by creating a data set and a benchmark for MER. The data set that we release, a MediaEval Database for Emotional Analysis in Music (DEAM), is the largest available data set of dynamic annotations (valence and arousal annotations for 1,802 songs and song excerpts licensed under Creative Commons with 2Hz time resolution). Using DEAM, we organized the ‘Emotion in Music’ task at MediaEval Multimedia Evaluation Campaign from 2013 to 2015. The benchmark attracted, in total, 21 active teams to participate in the challenge. We analyze the results of the benchmark: the winning algorithms and feature-sets. We also describe the design of the benchmark, the evaluation procedures and the data cleaning and transformations that we suggest. The results from the benchmark suggest that the recurrent neural network based approaches combined with large feature-sets work best for dynamic MER. [ABSTRACT FROM AUTHOR]

Knowledge-based neural network modeling techniques using space-mapping concept have been demonstrated in the existing literature as efficient methods to overcome the accuracy limitations of empirical/equivalent circuit models when matching new electromagnetic data. For different modeling problems, the mapping structures can be different. In this paper, we propose a unified automated model generation algorithm that uses \boldsymbol l\mathbf {1} optimization to automatically determine the type and the topology of the mapping structure in a knowledge-based neural network model. By encompassing various types of mappings of the knowledge-based neural network model in the existing literature, we present a new unified model structure and derive new sensitivity formulas for the training of the unified model. The proposed \boldsymbol l\mathbf {1} formulation of modeling can force some weights of the mapping neural networks to zeros while leaving other weights as nonzeros. We utilize this feature to allow \boldsymbol l\mathbf {1} optimization to automatically determine which mapping is necessary and which mapping is unnecessary. Using the proposed \boldsymbol l\mathbf {1} optimization method, the mapping structure can be determined to address different needs of different modeling problems. The structure of the final knowledge-based model can be flexible combinations of some or all of linear mapping, nonlinear mapping, input mapping, frequency mapping, and output mapping. In this way, the proposed algorithm is more systematic and can further speed up the knowledge-based modeling process than existing knowledge-based modeling algorithms. The proposed method is illustrated by three microwave filter modeling examples. [ABSTRACT FROM PUBLISHER]