检索结果-内蒙古大学图书馆

Research Square 2021年

作者： Zhao, Jian Zhang, Zhiwei Qiu, Jinping Shi, Lijuan Kuang, Zhejun Jing, Weipeng Zhu, Zhan Peng College of Computer Science and Technology Changchun University Changchun130022 China College of Electronic Information Engineering Changchun University Changchun130022 China Jilin Provincial Key Laboratory of Human Health Status Identification and Function Enhancement Changchun130022 China College of Information and Computer Engineering Northeast Forestry University Harbin China Department of Neurosurgery The First Hospital of Jilin University Jilin China

With the rapid development of deep learning in recent years, automatic electroencephalography (EEG) emotion recognition has been widely concerned. At present, most deep learning methods do not normalize EEG data properly and do not fully extract the features of time and frequency domain, which will affect the accuracy of EEG emotion recognition. To solve these problems, we propose GTScepeion, a deep learning EEG emotion recognition model. In pre-processing, the EEG time slicing data including channels were pre-processed. In our model, global convolution kernels are used to extract overall semantic features, followed by three kinds of temporal convolution kernels representing different emotional periods, followed by two kinds of spatial convolution kernels highlighting brain hemispheric differences to extract spatial features, and finally emotions are dichotomy classified by the full connected layer. The experiments is based on the DEAP dataset, and our model can effectively normalize the data and fully extract features. For Arousal, ours is 8.76% higher than the current optimal emotion recognition model based on Inception. For Valence, the best accuracy of our model reaches 91.51%. © 2021, CC BY.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Multi-Party dual learning

arXiv

引用

arXiv 2021年

作者： Gong, Maoguo Gao, Yuan Xie, Yu Qin, A.K. Pan, Ke Ong, Yew-Soon The School of Electronic Engineering Key Laboratory of Intelligent Perception and Image Understanding Ministry of Education Xidian University Xi'an Shaanxi Province710071 China The Key Laboratory of Computational Intelligence and Chinese Information Processing Ministry of Education Shanxi University Taiyuan030006 China The Department of Computer Science and Software Engineering Swinburne University of Technology Melbourne Australia The School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore

The performance of machine learning algorithms heavily relies on the availability of a large amount of training data. However, in reality, data usually reside in distributed parties such as different institutions and may not be directly gathered and integrated due to various data policy constraints. As a result, some parties may suffer from insufficient data available for training machine learning models. In this paper, we propose a multi-party dual learning (MPDL) framework to alleviate the problem of limited data with poor quality in an isolated party. Since the knowledge sharing processes for multiple parties always emerge in dual forms, we show that dual learning is naturally suitable to handle the challenge of missing data, and explicitly exploits the probabilistic correlation and structural relationship between dual tasks to regularize the training process. We introduce a feature-oriented differential privacy with mathematical proof, in order to avoid possible privacy leakage of raw features in the dual inference process. The approach requires minimal modifications to the existing multi-party learning structure, and each party can build flexible and powerful models separately, whose accuracy is no less than non-distributed self-learning approaches. The MPDL framework achieves significant improvement compared with state-of-the-art multi-party learning methods, as we demonstrated through simulations on real-world datasets. Copyright © 2021, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

All-Optical Signal processing of Vortex Beams with Diffractive Deep Neural Networks

引用

Physical Review Applied 2021年第1期15卷 014037-014037页

作者： Zebin Huang Peipei Wang Junmin Liu Wenjie Xiong Yanliang He Jiangnan Xiao Huapeng Ye Ying Li Shuqing Chen Dianyuan Fan International Collaborative Laboratory of 2D Materials for Optoelectronics Science and Technology of Ministry of Education Institute of Microscale Optoelectronics Shenzhen University Shenzhen 518060 China College of New Materials and New Energies Shenzhen Technology University Shenzhen 518118 China Terahertz Technology Innovation Research Institute Shanghai Key Lab of Modern Optical System Terahertz Science Cooperative Innovation Center School of Optical-Electrical Computer Engineering University of Shanghai for Science and Technology Shanghai 200093 China Guangdong Provincial Key Laboratory of Optical Information Materials and Technology & Institute of Electronic Paper Displays South China Academy of Advanced Optoelectronics South China Normal University Guangzhou 510006 China

Vortex beams (VBs), possessing a helical phase front and carrying orbital angular momentum (OAM), have attracted considerable attention in optical communications for their mode orthogonality. A platform for achieving all-optical signal processing of VBs, however, remains elusive due to the limited light-field-manipulation capability. We introduce diffractive deep neural networks (D2NNs) and their applications to process VBs. Exploiting the multiple-light-field-modulation ability of multilayer diffraction structures and the strong data-processing capability of deep neural networks, we reveal that D2NNs can manipulate multiple VBs by configuring the phase and amplitude distribution of diffractive screens. The diffraction efficiency and converted-mode purity are greater than 96%. After being trained, D2NNs with functions of hybrid-OAM-mode generation, identification, and conversion are obtained, and three typical types of all-optical signal-processing communication, (OAM-shift keying (OAM-SK), OAM multiplexing and demultiplexing, and OAM-mode switching) are successfully achieved. Our simulation results provide an approach that breaks the limitations of poor functionality and complex design in processing VBs, introducing the D2NN as a universal light-field-modulation platform.

关键词： Interference & diffraction of light Optical quantum information processing Optical vortices Quantum communication Artificial neural networks Multilayer thin films Machine learning Optical tweezers

来源：评论

学校读者我要写书评

暂无评论

A fair (t, n)-threshold secret sharing scheme with efficient cheater identifying 13th

A fair (t, n)-threshold secret sharing scheme with efficient...

引用

13th IFIP WG 11.11 International Conference on Trust Management, IFIPTM 2019

作者： Shen, Hua Sun, Daijie Zhao, Lan Zhang, Mingwu School of Computer Science Hubei University of Technology Wuhan China Hubei Key Laboratory of Intelligent Geo-Information Processing China University of Geosciences Wuhan China

ISBN: (纸本)9783030337155

The fairness of secret sharing guarantees that, if either participant obtains the secret, other participants obtain too. The fairness can be threatened by cheaters who was hidden in the participants. To efficiently and accurately identify cheaters with guaranteeing fairness, this paper proposes a fair (t, n)-threshold secret sharing scheme with an efficient cheater identifying ability. The scheme consists of three protocols which correspond to the secret distribution phase, secret reconstruction phase, and cheater identification phase respectively. The scheme’s secret distribution strategy enables the secret reconstruction protocol to detect the occurrence of cheating and trigger the execution of the cheater identification protocol to accurately locate cheaters. Moreover, we prove that the scheme is fair and secure, and show that the cheater identification algorithm has higher efficiency by comparing with other schemes. © IFIP International Federation for information processing 2019.

关键词： Security of data

来源：评论

学校读者我要写书评

暂无评论

Motif Graph Neural Network

arXiv

引用

arXiv 2021年

作者： Chen, Xuexin Cai, Ruichu Fang, Yuan Wu, Min Li, Zijian Hao, Zhifeng The School of Computer Science Guangdong University of Technology Guangzhou510006 China The School of Computer Science Guangdong University of Technology Guangdong Provincial Key Laboratory of Public Finance and Taxation with Big Data Application Guangzhou China Peng Cheng Laboratory Shenzhen518066 China The School of Computing and Information Systems Singapore Management University 178902 Singapore A*STAR 138632 Singapore The College of Science Shantou University Shantou515063 China

Graphs can model complicated interactions between entities, which naturally emerge in many important applications. These applications can often be cast into standard graph learning tasks, in which a crucial step is to learn low-dimensional graph representations. Graph neural networks (GNNs) are currently the most popular model in graph embedding approaches. However, standard GNNs in the neighborhood aggregation paradigm suffer from limited discriminative power in distinguishing high-order graph structures as opposed to low-order structures. To capture high-order structures, researchers have resorted to motifs and developed motif-based GNNs. However, existing motif-based GNNs still often suffer from less discriminative power on high-order structures. To overcome the above limitations, we propose Motif Graph Neural Network (MGNN), a novel framework to better capture high-order structures, hinging on our proposed motif redundancy minimization operator and injective motif combination. First, MGNN produces a set of node representations w.r.t. each motif. The next phase is our proposed redundancy minimization among motifs which compares the motifs with each other and distills the features unique to each motif. Finally, MGNN performs the updating of node representations by combining multiple representations from different motifs. In particular, to enhance the discriminative power, MGNN utilizes an injective function to combine the representations w.r.t. different motifs. We further show that our proposed architecture increases the expressive power of GNNs with a theoretical analysis. We demonstrate that MGNN outperforms state-of-the-art methods on seven public benchmarks on both node classification and graph classification tasks. Copyright © 2021, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

DLIENet: A lightweight low-light image enhancement network via knowledge distillation

引用

Pattern Recognition 2026年 169卷

作者： Ling Zhang Zhenyu Li Liang Cheng Qing Zhang Zheng Liu Xiaolong Zhang Chunxia Xiao School of Computer Science and Technology Wuhan University of Science and Technology Wuhan 430081 China Hubei Key Laboratory of Intelligent Information Processing and Realtime Industrial System Wuhan University of Science and Technology Wuhan 430081 China School of Computer Science and Engineering Sun Yat-Sen University Guangzhou 510275 China School of Computer Science China University of Geosciences (Wuhan) Wuhan 430074 China School of Computer Science Wuhan University Wuhan 430072 China

Learning-based methods have recently made remarkable progress for low-light image enhancement (LIE). However, most of these methods generally suffer from unsatisfactory results with overexposure and noise residual. To address these issues, we propose a distillation-based low-light image enhancement network (DLIENet), where we construct a teacher network for LIE to help train DLIENet with lightweight architecture. DLIENet imitates the low-light enhancement capability of the teacher network through knowledge distillation. Concretely, the distillation in the bottleneck module transfers the feature representation capability from the pre-trained teacher network to DLIENet. The distillation in content attention module forces DLIENet to focus on the key contents for reconstructing normal-light images, promoting noise-free results for the low-light images. Additionally, we employ a two-step joint training strategy to improve the performance of the teacher network for LIE. Moreover, we introduce a feature fusion block in DLIENet to normalize the features globally, avoiding non-uniform illumination and overexposure. Extensive experiments and evaluations on both the synthetic and real-world datasets demonstrate the generalization and generalization ability of our method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Syntax-Aware Chinese Frame Semantic Role Labeling Based on Self-Attention 19

Syntax-Aware Chinese Frame Semantic Role Labeling Based on S...

引用

19th Chinese National Conference on Computational Linguistic, CCL 2020

作者： Wang, Xiaohui Li, Ru Wang, Zhiqiang Chai, Qinghua Han, Xiaoqi School of Computer and Information Technology Shanxi University China Key Laboratory of Ministry of Education for Computational Intelligence and Chinese Information Processing Shanxi University China Institute of Intelligent Information Processing Shanxi University China School of Foreign Languages Shanxi University China

Frame semantic role labeling is a semantic analysis task based on the FrameNet. Semantic role labeling usually relies heavily on syntax. Most of the existing semantic role labeling models are based on Bi-LSTM. Although they can obtain the long-distance dependency information in sentences, they can not get the syntactic information well. Therefore, the self attention mechanism is introduced to capture the syntactic information of each word in the sentence. Experimental results show that the Fl of our model on Chinese FrameNet dataset reaches 83.77%, which is increased by nearly 11%. © 2020 China National Conference on Computational Linguistics Published under Creative Commons Attribution 4.0 International License

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Color image segmentation based on a convex K-means approach

arXiv

引用

arXiv 2021年

作者： Wu, Tingting Gu, Xiaoyu Shao, Jinbo Zhou, Ruoxuan Li, Zhi School of Science Nanjing University of Posts and Telecommunications Nanjing210023 China College of Oceanography and Space Informatics China University of Petroleum Qingdao266580 China Department of Computer Science and Technology Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai200241 China

Image segmentation is a fundamental and challenging task in image processing and computer vision. The color image segmentation is attracting more attention due to the color image provides more information than the gray image. In this paper, we propose a variational model based on a convex K-means approach to segment color images. The proposed variational method uses a combination of l1 and l2 regularizers to maintain edge information of objects in images while overcoming the staircase effect. Meanwhile, our onestage strategy is an improved version based on the smoothing and thresholding strategy, which contributes to improving the accuracy of segmentation. The proposed method performs the following steps. First, we specify the color set which can be determined by human or the K-means method. Second, we use a variational model to obtain the most appropriate color for each pixel from the color set via convex relaxation and lifting. The Chambolle-Pock algorithm and simplex projection are applied to solve the variational model effectively. Experimental results and comparison analysis demonstrate the effectiveness and robustness of our method. © 2021, CC BY.

关键词： Relaxation processes

来源：评论

学校读者我要写书评

暂无评论

A new classification method of SAR image based on OGMRF-RC model 5

A new classification method of SAR image based on OGMRF-RC m...

引用

IET International Radar Conference 2020, IET IRC 2020

作者： Li, Yuan Guo, Zhishun Wu, Lin Guo, Zhengwei College of Computer and Information Engineering Henan University Kaifeng475004 China Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng475004 China Henan Engineering Research Center of Intelligent Technology and Application Henan University Kaifeng475000 China College of Environment and Planning Henan University Kaifeng475004 China

ISBN: (纸本)9781839535635

Synthetic Aperture Radar (SAR) image classification is a key technique and research focus in the application of remote sensing. This paper proposes a new classification method of SAR image based on Object-based Gaussian-Markov Random Field model with Region Coefficients (OGMRF-RC), the new method use Fuzzy Probability (FP) represents the region's category probability, FP is obtained by regional posterior probability and edge information of neighborhood. Use FP to replace the unique class label of the region, which avoids the parameters estimation inaccurate of the feature field caused by the unique initial class label. And the parameter calculation method of feature field takes into account the feature information of all regions in the image. In this paper, we take the area around Baliwan in Kaifeng as study area, Sentinel-1 is used as the remote sensing data sources. FCM, K-means, classic MRF and OGMRF-RC are used as contrast methods, the experimental result demonstrates the accuracy of our method is better than these methods. © 2020 Institution of Engineering and technology. All rights reserved.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

Age-of-Task Aware Sampling Rate Optimization in Edge-Assisted Industrial Network Systems

Age-of-Task Aware Sampling Rate Optimization in Edge-Assiste...

引用

2021 IEEE Global Communications Conference (GLOBECOM)

作者： Xiaojing Wen Cailian Chen Mingyan Li Ling Lyu Xinping Guan Key Laboratory of System Control and Information Processing Shanghai Jiao Tong University Ministry of Education of China Shanghai P.R. China College of Computer Science Chongqing University Chongqing P. R. China School of Information Science and Technology Dalian Maritime University Dalian P. R. China

ISBN: (纸本)9781728181059

Multivariate-information and computation-intensive tasks play an important role in Wireless Sensor Network Systems (WSNSs), where information freshness has an important impact on the system performance of state analysis. Recently, the Age of information (AoI) has been studied extensively as a promising metric to evaluate the freshness of state packets. However, most of the existing research focuses on optimizing the average AoI of a single information source, which can not be directly applied to the scenario with multi-source tasks. In this paper, we firstly present a novel definition of the Age of Task (AoT) for edge-assisted industrial WSNSs. Furthermore, the expressions of the sensing time and arrival time that determine the AoT tail distribution are given in detail. Then, we propose an AoT tail violation probability minimization problem to find the optimal sampling rate and give the feasible region of the sampling rate. Since it is difficult to obtain the exacted expression of the formulated problem, the Upper Bound Minimization Problem (UBMP) and the more tractable α-relaxed UBMP are proposed to obtain the near-optimal sampling rate. Finally, simulation results show that the sampling rate obtained by α-relaxed UBMP is nearly optimal for the AoT tail violation probability minimization problem.

关键词： Wireless sensor networks Analytical models Upper bound System performance Simulation Stochastic processes Minimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：