scispace - formally typeset
Search or ask a question

Showing papers in "Neurocomputing in 2021"


Journal ArticleDOI
TL;DR: An overview of the state-of-the-art attention models proposed in recent years is given and a unified model that is suitable for most attention structures is defined.

620 citations


Journal ArticleDOI
TL;DR: A comprehensive review of the recent developments on deep face recognition can be found in this paper, covering broad topics on algorithm designs, databases, protocols, and application scenes, as well as the technical challenges and several promising directions.

353 citations


Journal ArticleDOI
TL;DR: A survey on two types of network compression: pruning and quantization is provided, which compare current techniques, analyze their strengths and weaknesses, provide guidance for compressing networks, and discuss possible future compression techniques.

266 citations


Journal ArticleDOI
TL;DR: Online learning as mentioned in this paper is a family of machine learning methods, where a learner attempts to tackle some predictive (or any type of decision-making) task by learning from a sequence of data instances one by one at each time.

234 citations


Journal ArticleDOI
TL;DR: Results for every optimization task demonstrate that LSEOFOA can provide a high-performance and self-assured tradeoff between exploration and exploitation, and overall research findings show that the proposed model is superior in terms of classification accuracy, Matthews correlation coefficient, sensitivity, and specificity.

212 citations


Journal ArticleDOI
TL;DR: Four main methods of transfer learning are described and their practical applications in EEG signal analysis in recent years are explored.

184 citations


Journal ArticleDOI
TL;DR: Although there is a vast gap between deep learning models (that usually need sufficient labeled samples) and the HSI scenario with few labeled samples, the issues of small-sample sets can be well characterized by fusion of deep learning methods and related techniques, such as transfer learning and a lightweight model.

170 citations


Journal ArticleDOI
TL;DR: This paper proposes a template update mechanism to improve the accuracy of visual tracking and shows that the proposed mechanism has improved accuracy and success rate of the two baseline algorithms and in state-of-the-art algorithms.

137 citations


Journal ArticleDOI
TL;DR: A novel image segmentation method is developed in this paper for quantitative analysis of GICS based on the deep reinforcement learning (DRL), which can accurately distinguish the test line and the control line in the GICS images.

119 citations


Journal ArticleDOI
TL;DR: In this article, a detailed analysis of the influence of non-IID data on both parametric and non-parametric machine learning models in both horizontal and vertical federated learning is provided.

117 citations


Journal ArticleDOI
TL;DR: A novel meta-learning fault diagnosis method (MLFD) based on model-agnostic meta- learning that achieves fast and accurate few-shot bearing fault diagnosis under unseen working conditions by leveraging the learned knowledge.

Journal ArticleDOI
TL;DR: An overview of the use of CNNs, for image classification, segmentation, detection, and other tasks such as registration, content-based image retrieval, image generation and enhancement, in some typical medical diagnosis areas such as brain, breast, and abdominal are presented.

Journal ArticleDOI
TL;DR: A new method of data missing estimation with tensor heterogeneous ensemble learning based on FNN (Fuzzy Neural Network) named FNNTEL is proposed in this paper and the performance is better than other commonly used technologies and different missing data generation models.

Journal ArticleDOI
TL;DR: A model that has joint weak saliency and attention aware is proposed, which can obtain more complete global features by weakening saliency features and obtains diversifiedsaliency features via attention diversity to improve the performance of the model.

Journal ArticleDOI
TL;DR: The most effective techniques emerging within this branch of research are presented to identify remaining challenges as well as to build upon this platform of work towards further novel techniques for handling irregular time series data.

Journal ArticleDOI
TL;DR: Recently, a large body of deep learning methods have been proposed and has shown great promise in handling the traditional ill-posed problem of depth estimation as discussed by the authors, which is of great significance for many applications such as augmented reality, target tracking and autonomous driving.

Journal ArticleDOI
TL;DR: Experimental results demonstrate that the developed AADL-based labels have several advantages, such as robustness for head pose image missing, insensitivity for the motion blur, and good performance compared to several state-of-the-art methods on the Pointing’04 and CAS_PEAL_R1 databases.

Journal ArticleDOI
TL;DR: In this article, a comprehensive review on deep multi-view learning from the following two perspectives: MVL methods in deep learning scope and deep MVL extensions of traditional methods is presented, and the authors attempt to identify some open challenges to inform future research directions.

Journal ArticleDOI
TL;DR: An integrated method of deep learning for prediction of time series that incorporates network models including both bi-directional and grid long short-term memory network to achieve high-quality prediction of workload and resource time series is designed.

Journal ArticleDOI
TL;DR: This paper formulate causality extraction as a sequence labeling problem based on a novel causality tagging scheme, and proposes a neural causality extractor with the BiLSTM-CRF model as the backbone, named SCITE (Self-attentive BiL STM- CRF wIth Transferred Embeddings), which can directly extract cause and effect without extracting candidate causal pairs and identifying their relations separately.

Journal ArticleDOI
TL;DR: This survey focuses on local interpretation methods of deep neural networks with an in-depth analysis of the representative works including the newly proposed approaches, and makes a fine-grained distinction between the two types of these methods.

Journal ArticleDOI
TL;DR: In this article, a supervised multi-label classification framework based on deep convolutional neural networks (CNNs) was proposed for predicting the presence of 14 common thoracic diseases and observations.

Journal ArticleDOI
TL;DR: A novel method of adversarial knowledge transfer named SA-GAN stands for Subject Adaptor GAN, which utilizes the Generative Adversarial Network framework to perform cross-subject transfer learning in the domain of wearable sensor-based Human Activity Recognition.

Journal ArticleDOI
TL;DR: The results show that, for most of the problems considered, X-TFC achieves high accuracy with low computational time, even for large scale PDEs, without suffering the curse of dimensionality.

Journal ArticleDOI
TL;DR: A novel detection method called self-regularized weighted sparse (SRWS) model, designed for the hypothesis that data may come from multi-subspaces is proposed, which outperforms state-of-the-art baselines and optimized its iterative convergence condition.

Journal ArticleDOI
TL;DR: Both delay-independent and delay-dependent criteria to guarantee the existence, uniqueness and global stability of equilibrium point for the considered FOQVNNs are derived in the form of linear matrix inequality (LMI).

Journal ArticleDOI
TL;DR: All the algorithms studied in this paper will be evaluated with exhaustive testing in order to analyze their capabilities in standard classification problems, particularly considering dimensionality reduction and kernelization.

Journal ArticleDOI
TL;DR: A survey of dynamic network embedding can be found in this paper, where the authors inspect the data model, representation learning technique, evaluation and application of current related works and derive common patterns from them.

Journal ArticleDOI
TL;DR: A novel model based on 3D fully convolutional network is proposed that applies multi-pathway architecture to feature extraction so as to effectively extract features from multi-modal MRI images.

Journal ArticleDOI
Le Yu1, Bowen Du1, Xiao Hu1, Leilei Sun1, Liangzhe Han1, Weifeng Lv1 
TL;DR: Experimental results on real-world datasets demonstrate that DSTGCN outperforms both classical and state-of-the-art methods to predict traffic accidents.