Showing papers in "Neurocomputing in 2019"

TL;DR: A novel and unified architecture which contains a bidirectional LSTM (BiLSTM), attention mechanism and the convolutional layer is proposed in this paper, which outperforms other state-of-the-art text classification methods in terms of the classification accuracy.

TL;DR: A deep learning approach capable to address the limitations of traditional forecasting approaches and show accurate predictions is proposed, and the empirical results show that the proposed DLSTM model outperforms other standard approaches.

TL;DR: The three popular Deep Learning algorithms for Bearing fault diagnosis including Autoencoder, Restricted Boltzmann Machine, and Convolutional Neural Network are briefly introduced and their applications are reviewed through publications and research works on the area of bearing fault diagnosis.

TL;DR: A survey of semantic segmentation methods by categorizing them into ten different classes according to the common concepts underlying their architectures, and providing an overview of the publicly available datasets on which they have been assessed.

TL;DR: A deep hierarchical convolutional neural network (CNN) is proposed, called as DeepCrack, to predict pixel-wise crack segmentation in an end-to-end method using both guided filtering and Conditional Random Fields methods to refine the final prediction results.

TL;DR: In this article, a test-time augmentation-based aleatoric uncertainty was proposed to analyze the effect of different transformations of the input image on the segmentation output, and the results showed that the proposed test augmentation provides a better uncertainty estimation than calculating the testtime dropout-based model uncertainty alone and helps to reduce overconfident incorrect predictions.

TL;DR: A novel capsule network with an inception block and a regression branch is proposed, Inspired by dynamic routing capsule net, for Bearing fault diagnosis of rotating machine health monitoring.

TL;DR: A thorough review of existing types of image steganography and the recent contributions in each category in multiple modalities including general operation, requirements, different aspects, different types and their performance evaluations is provided.

TL;DR: A novel Deep Residual Reconstruction Network (DR2-Net) to reconstruct the image from its Compressively Sensed measurement by outperforms traditional iterative methods and recent deep learning-based methods by large margins at measurement rates 0.01, 0.1, and 0.25.

TL;DR: In this paper, the authors present a taxonomy of CNN acceleration methods in terms of three levels, i.e. structure level, algorithm level, and implementation level, for CNN architectures compression, algorithm optimization and hardware-based improvement.

TL;DR: This work proposes an improved approach that connects the high-impact value of remarkably long sequence time steps to the current time step, and these high- impact traffic flow values are captured using the attention mechanism.

TL;DR: This work proposes a deep learning based approach for supervised multi-time series anomaly detection that combines a Convolutional Neural Network and a Recurrent Neural Network in different ways and refers to this architecture as Multi-head CNN–RNN.

TL;DR: Evaluated CNN-based architectures on three T2-weighted MRI datasets, each one consisting of a different number of patients and heterogeneous image characteristics, collected by different institutions show that training on the union of the datasets generally outperforms training on each dataset separately, allowing for both intra-/cross-dataset generalization.

TL;DR: The experimental results show that the proposed method can not only solve the two deficiencies of SAEs, but also achieve a superior performance to the existing methods.

TL;DR: This survey focuses on recent DL approaches that have been proposed in the area of cybersecurity, namely intrusion detection, malware detection, phishing/spam detection, and website defacement detection.

TL;DR: A new objective function that is suitable for the RUL estimation problem is proposed, as well as a new target generation approach for training LSTM networks, which requires making lesser assumptions about the actual degradation of the system.

TL;DR: The resulting CNN is shown to provide better classification performance when compared to more conventional learning machines; indeed, it achieves an average accuracy of 89.8% in binary classification and of 83.3% in three-ways classification.

TL;DR: An improved CNN named multi-scale cascade convolutional neural network (MC-CNN) is proposed for the classification information enhancement of input and is verified by analyzing the application of MC-CNN in bearing fault diagnosis under nonstationary working conditions.

TL;DR: This paper introduces the Spatial-/Channel-wise Attention Models into the traditional Regression CNN to estimate the density map, which is named as “SCAR” and achieves state-of-the-art results.

TL;DR: In this article, the authors proposed a new Autoregressive integrated moving average (ARIMA)-Artificial Neural Network (ANN) hybrid method that work in a more general framework.

TL;DR: This paper surveys the literature on security aspects of CPSs, and presents some of existing methods for detecting cyber attacks, which are: Denial of service (DoS), deception, and replay attacks.

TL;DR: An individual brain network is constructed as feature representation, and a deep neural network (DNN) classifier is used to perform ASD/TC classification via a DNN classifier.

TL;DR: A comprehensive review of current six types of deep learning methods designed for person re-identification, and the detailed descriptions of existing person ReID datasets are presented.

TL;DR: Experimental results show that the proposed method has outstanding performance among other excellent approaches on identifying drug-side effect associations, and compared with many existing methods, the proposed approach achieves better results on three benchmark datasets of drug side-effect associations.

TL;DR: This paper is concerned with the finite-time and the fixed-time synchronization problem for a class of inertial neural networks with multi-proportional delays and some new and effective criteria are established to achieve finite- time and fixed- time synchronization of the master/slave of addressed systems.

TL;DR: This chapter proposed an ensemble of 3D densely connected convolutional networks for AD and MCI diagnosis from 3D MRIs and superior performance of the proposed model was demonstrated on ADNI dataset.

TL;DR: A robust and fixed-time zeroing neural dynamics model is proposed and analyzed for time-variant nonlinear equation (TVNE), and comparative results demonstrate the effectiveness, robustness, and advantage of the RaFT-ZND model for solving TVNE.

TL;DR: A high-dimensional mixed integer nonlinear layout optimization mathematical model involving the pipeline network structure parameters and pipeline design parameters is established, which can be applied to large-scale oil gathering system and gas gathering system universally.

TL;DR: A hyperspectral image (HSI) classification method using spectral-spatial long short term memory (LSTM) networks, which can improve the classification accuracy by at least 2.69%, 1.53% and 1.08% compared to other state-of-the-art methods.

TL;DR: A comprehensive review of the significant studies exploited Artificial Neural Networks (ANNs) in BEA (Building Energy Analysis) with a three-decade time span of the publishing date of the existing studies was taken into account.