检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wu, Zhiyuan Sun, Sheng Wang, Yuwei Liu, Min Xu, Ke Pan, Quyang Gao, Bo Wen, Tian The State Key Lab of Processers Institute of Computing Technology Chinese Academy of Sciences Beijing China The University of Chinese Academy of Sciences Beijing China The Zhongguancun Laboratory Beijing China The Department of Computer Science and Technology Tsinghua University Beijing China The School of Computer Science and Technology The Engineering Research Center of Network Management Technology for High-Speed Railway of Ministry of Education Beijing Jiaotong University Beijing China

The rise of End-Edge-Cloud Collaboration (EECC) offers a promising paradigm for Artificial Intelligence (AI) model training across end devices, edge servers, and cloud data centers, providing enhanced reliability and reduced latency. Hierarchical Federated Learning (HFL) can benefit from this paradigm by enabling multi-tier model aggregation across distributed computing nodes. However, the potential of HFL is significantly constrained by the inherent heterogeneity and dynamic characteristics of EECC environments. Specifically, the uniform model structure bounded by the least powerful end device across all computing nodes imposes a performance bottleneck. Meanwhile, coupled heterogeneity in data distributions and resource capabilities across tiers disrupts hierarchical knowledge transfer, leading to biased updates and degraded performance. Furthermore, the mobility and fluctuating connectivity of computing nodes in EECC environments introduce complexities in dynamic node migration, further compromising the robustness of the training process. To address multiple challenges within a unified framework, we propose End-Edge-Cloud Federated Learning with Self-Rectified Knowledge Agglomeration (FedEEC), which is a novel EECC-empowered FL framework that allows the trained models from end, edge, to cloud to grow larger in size and stronger in generalization ability. FedEEC introduces two key innovations: (1) Bridge Sample Based Online Distillation Protocol (BSBODP), which enables knowledge transfer between neighboring nodes through generated bridge samples, and (2) Self-Knowledge Rectification (SKR), which refines the transferred knowledge to prevent suboptimal cloud model optimization. The proposed framework effectively handles both cross-tier resource heterogeneity and effective knowledge transfer between neighboring nodes, while satisfying the migration-resilient requirements of EECC. Extensive experiments on three datasets demonstrate that FedEEC achieves significantly hi

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

data-driven Quantum Dynamical Embedding Method for Long-term Prediction on Near-term Quantum Computers

arXiv

引用

arXiv 2023年

作者： Sun, Tai-Ping Chen, Zhao-Yun Xue, Cheng Liu, Huan-Yu Zhuang, Xi-Ning Wang, Yun-Jie Ma, Shi-Xin Zhang, Hai-Feng Wu, Yu-Chun Guo, Guo-Ping Key Laboratory of Quantum Information Chinese Academy of Sciences School of Physics University of Science and Technology of China Anhui Hefei230026 China CAS Center For Excellence in Quantum Information and Quantum Physics University of Science and Technology of China Anhui Hefei230026 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Anhui Hefei230088 China Origin Quantum Computing Anhui Hefei230026 China Institute of Advanced Technology University of Science and Technology of China Anhui Hefei230031 China National Innovation Institute of Defense Technology AMS Beijing100071 China

The increasing focus on long-term time series prediction across various fields has been significantly strengthened by advancements in quantum computation. In this paper, we introduce a data-driven method designed for long-term time series prediction with quantum dynamical embedding (QDE). This approach enables a trainable embedding of the data space into an extended state space, allowing for the recursive retrieval of time series information. Based on its independence of time series length, this method achieves depth-efficient quantum circuits that are crucial for near-term quantum computers. Numerical simulations demonstrate the model’s improved performance in prediction accuracy and resource efficiency over existing methods, as well as its effective denoising capabilities. We implement this model on the Origin "Wukong" superconducting quantum processor with a learnable error-cancellation layer (LECL) for error mitigation, further validating the practical applicability of our approach on near-term quantum devices. Furthermore, the theoretical analysis of the QDE’s dynamical properties and its universality enhances its potential for time series prediction. This study establishes a significant step towards the processing of long-term time series on near-term quantum computers, integrating data-driven learning with discrete dynamical embedding for enhanced forecasting capabilities. © 2023, CC0.

关键词： Quantum computers

来源：评论

学校读者我要写书评

暂无评论

A Multi-DOF Robot System Based on LightGBM-Driven EEG Decoding Model for BCI Human-Machine Interaction

A Multi-DOF Robot System Based on LightGBM-Driven EEG Decodi...

引用

Chinese Control Conference (CCC)

作者： Weidong Yan Zhaoliang Xu Yang Li The Department of Automation Science and Electrical Engineering Beihang University Beijing China The Beijing Advanced Innovation Center for Big Data and Brain Computing Beijing China State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China Advanced Institute of Information Technology Peking University Hangzhou China

In this study, a multi-degree-of-freedom (Multi-DOF) robot (MDR) system based on a LightGBM-driven electroencephalogram (EEG) decoding model is designed and developed to assist subjects with hand motor dysfunction in their daily activities and neurorehabilitation. The system mainly consists of a motor imagery electroencephalogram (MI-EEG) evoking layer, an intention decoding layer, and an interaction executive layer. The MI-EEG evoking layer initially displays a virtual reality (VR) motion imagination scenario, instructing the subjects to imagine a real hand gripping movement, simultaneously collecting the electrical EEG signals and preprocessing the EEG signals. Secondly, in the intention decoding layer, a network combining temporal-spectral feature fusion and LightGBM (TSFF-LightGBM) for MI-BCI classification is used to more effectively boost brain decoding accuracy and decrease decoding time. Finally, in the interaction executive layer, the Multi-DOF wearable robot is developed to offer hand grasp motion kinesthetic feedback and visual feedback synced with MI. The following are the key benefits of the proposed MDR system: (1) We propose a new lightweight network structure more suitable for brain computer interface (BCI) interaction systems, achieving more accurate decoding and shorter identification time in different data sets, which helps to improve the practicability of the system and promote the practical clinical application of BCI rehabilitation technology. (2) We integrate BCI, VR, a wearable Multi-DOF robot, motion kinesthetic feedback, and visual feedback to improve human-machine interaction. Compared to the most recent investigations, the average accuracy of the MDR system on the publicly accessible datasets BCI IV 2a and HGD reached 75.89% and 93.53%, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Field Test of Twin-Field Quantum key Distribution through Sending-or-Not-Sending over 428 km

引用

Physical Review Letters 2021年第25期126卷 250502-250502页

作者： Hui Liu Cong Jiang Hao-Tao Zhu Mi Zou Zong-Wen Yu Xiao-Long Hu Hai Xu Shizhao Ma Zhiyong Han Jiu-Peng Chen Yunqi Dai Shi-Biao Tang Weijun Zhang Hao Li Lixing You Zhen Wang Yong Hua Hongkun Hu Hongbo Zhang Fei Zhou Qiang Zhang Xiang-Bin Wang Teng-Yun Chen Jian-Wei Pan Hefei National Laboratory for Physical Sciences at Microscale and Department of Modern Physics University of Science and Technology ofChina Hefei Anhui 230026 People’s Republic of China CAS Center for Excellence in Quantum Information and Quantum Physics University of Science and Technology of China Hefei Anhui 230026 People’s Republic of China Jinan Institute of Quantum Technology Jinan Shandong 250101 People’s Republic of China State Key Laboratory of Low Dimensional Quantum Physics Department of Physics Tsinghua University Beijing 100084 People’s Republic of China Data Communication Science and Technology Research Institute Beijing 100191 People’s Republic of China QuantumCTek Corporation Limited Hefei Anhui 230088 People’s Republic of China State Key Laboratory of Functional Materials for Informatics Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Shanghai 200050 People’s Republic of China Chongqing Optoelectronics Research Institute Chongqing 400060 People’s Republic of China

Quantum key distribution endows people with information-theoretical security in communications. Twin-field quantum key distribution (TF-QKD) has attracted considerable attention because of its outstanding key rates over long distances. Recently, several demonstrations of TF-QKD have been realized. Nevertheless, those experiments are implemented in the laboratory, and therefore a critical question remains about whether the TF-QKD is feasible in real-world circumstances. Here, by adopting the sending-or-not-sending twin-field QKD (SNS-TF-QKD) with the method of actively odd parity pairing (AOPP), we demonstrate a field-test QKD over 428 km of deployed commercial fiber and two users are physically separated by about 300 km in a straight line. To this end, we explicitly measure the relevant properties of the deployed fiber and develop a carefully designed system with high stability. The secure key rate we achieved breaks the absolute key rate limit of repeaterless QKD. The result provides a new distance record for the field test of both TF-QKD and all types of fiber-based QKD systems. Our work bridges the gap of QKD between laboratory demonstrations and practical applications and paves the way for an intercity QKD network with measurement-device-independent security.

关键词： Optical quantum information processing Quantum cryptography Quantum optics

来源：评论

学校读者我要写书评

暂无评论

Multi-Channel Hypergraph network for Sequential Diagnosis Prediction in Healthcare

Multi-Channel Hypergraph Network for Sequential Diagnosis Pr...

引用

International Conference on Computer Supported Cooperative Work in Design

作者： Xin Zhang Xueping Peng Weimin Chen Weiyu Zhang Xiaoqiang Ren Wenpeng Luv Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Australian Artificial Intelligence Institute University of Technology Sydney Sydney Australia Shandong Mental Health Center Jinan China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Sequential diagnosis prediction (SDP) is a complex and challenging task, aming to predict future diagnoses of patients by analyzing their historical medical records. Although graph neural networks(GNNs) has been applied to successfully address the challenge of heterogeneous data integration in electronic health records, relatively limited work has been done on GNNs for sequential diagnosis prediction. Graph neural network-based methods, aimed at capturing structural and relational patterns of EHR data for sequential diagnosis prediction, explore code-code pairwise relationships, resulting in an inability to learn fine-grained, higher-order interaction relationships among different types medical codes. As a result, they are difficult to effectively model complex, multi-dimensional interactions among different types of medical codes necessary for accurate and nuanced diagnosis predictions. To address these challenges, this paper proposes a novel approach called Multi-Channel Hypergraph network (MCHN) predictive framework for sequential diagnosis prediction. The proposed method aims to explore the fine-grained higher-order interactions between different types of medical codes via multi-channel hypergraphs. Specifically, MCHN learns two levels of code embeddings from multi-channel hypergraph learning module and line graph learning module, respectively: (i) multi-channel hypergraph learning module, which is to learn multi-channel hypergraph level code embeddings by modeling the higher-order relationships between medical codes in different hypergraphs; and (ii) line graph learning module, which is to learn the line graph level code embeddings by modeling code-code pairwise relationships. In MCHN, we propose a novel channel-level attention mechanism to help our model attend to the informativeness of the different channel for forecasting future patient diagnoses. We also design a code-level attention mechanism, which can to pay more attention to the medical codes that are m

关键词： Codes Attention mechanisms Federated learning Medical services Predictive models Benchmark testing Sequential diagnosis

来源：评论

学校读者我要写书评

暂无评论

A Multi-Head Convolution network with Attention Consistency for Facial Expression Recognition

A Multi-Head Convolution Network with Attention Consistency ...

引用

Chinese Control Conference (CCC)

作者： Wenkang Liu Mingyi Sun Yang Li the Department of Automation Science and Electrical Engineering Beihang University Beijing China the Beijing Advanced Innovation Center for Big Data and Brain Computing Beijing China State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China Advanced Institute of Information Technology Peking University Hangzhou China

In recent years, the demand for facial expression recognition applications has increased rapidly, and its research has received extensive attention from researchers. However, the current recognition methods based on deep learning ignore the idea of multiple head attention and semantic consistency, resulting in the model only paying attention to the local area of the feature map. In addition, the model's inconsistent attention to the images before and after the flip, resulting the poor robustness, poor interpretation, and other shortcomings in the model. To address the above problems, we propose an Affinity Separation Loss (ASLoss), which improves the separability of samples through clustering. Moreover, a Separate Multi-head Attention block (SMA), and a Zonal Loss (ZLoss) are also designed to decentralize the model's attention. Experimental results demonstrate that our proposed MACNet method achieves competitive recognition performance on two public datasets RAF-DB and FERPlus.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design of Gesture Recognition System Based on Multi-Channel Myoelectricity Correlation

Design of Gesture Recognition System Based on Multi-Channel ...

引用

2019 IEEE Global Communications Conference (GLOBECOM)

作者： Di Wu Huiyong Li Xuefeng Liu Jianwei Niu State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Beijing Advanced Innovation Center for Big Data and Brain Computing (BDBC) Beihang University Hangzhou Innovation Research Institute Beihang University

ISBN: (数字)9781728109626

ISBN: (纸本)9781728109633

Gesture recognition systems based on myoelectric signal have raised more and more attention from researchers. Traditional gesture recognition methods are susceptible to multiple types of noise and require a large number of features, which increase overhead and decrease recognition efficiency. Fully utilizing the characteristics of the signal to recognize gestures is a big challenge. This paper proposes an improved empirical mode decomposition method (XB-EMD) based on autocorrelation function to denoise myoelectric signal. In addition, a novel deep neural network (CRNet) which combines the CNN and RNN layers together is trained for classifying the gestures based on denoised myoelectric signal. Experimental results show that the proposed gesture recognition system can improve recognition effectiveness at 97.4\% with 10 typical gestures, and 72.55\% with 16 complicated gestures.

关键词： Correlation Gesture recognition Feature extraction Empirical mode decomposition Recurrent neural networks Sensors

来源：评论

学校读者我要写书评

暂无评论

A Predictive Framework for Shipborne Wind Speed Measurement Correction Based on Self-Supervised Contrastive Learning

A Predictive Framework for Shipborne Wind Speed Measurement ...

引用

International Conference on Computer Supported Cooperative Work in Design

作者： Jian Song Xiang Li Zhenqiang Zhang Zhigang Zhao Chunxiao Wang Shunfang Wu Suiping Qi Jialiang Lv Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Key Laboratory of Computing Power Network and Information Security Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Institute of Oceanographic Instrumentation Qilu University of Technology (Shandong Academy of Sciences) Qingdao China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Accurate measurement of wind speed on maritime vessels is crucial for weather and sea condition forecasting, safe navigation, power generation, hydrological simulation, and other applications. However, the precision of measurements may be subject to certain errors due to factors such as vessel motion and environmental conditions. To enhance the precision of shipborne wind speed measurement, this paper introduces an innovative approach based on contrastive learning. Through proficient feature extraction and the application of a self-supervised contrastive learning algorithm, this method extracts features of varying granularity from marine observational data to predict and correct shipborne wind speed measurements. To the best of our knowledge, this study represents the first attempt to validate contrastive learning in the intelligent analysis of marine observational data. Validation experiment results demonstrate the outstanding performance of this method in both single-step and multi-step predictions, showcasing higher efficacy and robustness compared to alternative approaches.

关键词： Accuracy Power measurement Wind speed Time series analysis Sea measurements Contrastive learning Predictive models

来源：评论

学校读者我要写书评

暂无评论

A Semantic-based Replacement for Event Image Privacy

A Semantic-based Replacement for Event Image Privacy

引用

IEEE International Conference on Computational science and Engineering, CSE

作者： Zhenfei Chen Zhu Tianqing Bing Tian Yu Wang Wei Ren School of Computer Science China University of Geosciences Wuhan P.R. China Institute of Artificial Intelligence and Blockchain Guangzhou University Guangzhou P.R. China School of Computer Science China University of Geosciences Guangxi Key Laboratory of Cryptography and Information Security Key Laboratory of Network Assessment Technology CAS Beijing P.R. China

ISBN: (纸本)9781665416610

Thanks to the continuous development of deep learning and the updating of deep neural networks, the accuracy of various computer vision tasks continue improving. On the one hand, the accuracy of image recognition is significantly improved. On the other hand, it also poses a higher challenge of image-based privacy preservation. Although traditional privacy protection methods such as cryptography methods can provide a good privacy protection, they are extremely inconvenient to use and cannot provide good image utility. In order to obtain a balance between image privacy and utility, we propose a privacy-preserving model based on image semantic replacement. We perform semantic replacement or obfuscation to multiple information. Taking the human figures as an example, the information includes faces, scenes, and dressing style. As that information contributes the most to the recognition, we define those items as the privacy of the original image. We replace the event information of the original image, so that the figures in the image can no longer be recognized. With this strategy, the image can still be detected by various detection networks, such as scene detection, which ensures utility. The framework consists of three parts: detection network, scene replacement network, and clothing replacement network. A comprehensive and quantitative experiment set proves the effectiveness of the proposed model.

关键词： Deep learning Privacy Image recognition Scientific computing Semantics Clothing Neural networks

来源：评论

学校读者我要写书评

暂无评论

Prefrontal Internal Event-Driven Analysis of Dynamical Electroencephalographic Biomarkers in Depression During Emotional Auditory Task

SSRN

引用

SSRN 2024年

作者： Zhao, Qinglin Cui, Kunbo Jiang, Hua Wu, Zhongqing Zhang, Lixin Zhao, Mingqi Hu, Bin Gansu Provincial Key Laboratory of Wearable Computing School of Information Science and Engineering Lanzhou University Lanzhou730000 China School of Medical Technology Beijing Institute of Technology Beijing100081 China CAS Center for Excellence in Brain Science and Intelligence Technology Shanghai Institutes for Biological Sciences Chinese Academy of Sciences Shanghai200000 China Joint Research Center for Cognitive Neurosensor Technology Lanzhou University Institute of Semiconductors Chinese Academy of Sciences Lanzhou730000 China

The prefrontal cortex is considered as one of the key brain regions for the study of affective dysfunction in depression. The study of neural processes in this region during specific emotional tasks may enhance our understanding on the mechanisms of depression. One powerful, direct and convenient tool is electroencephalography (EEG);yet in many experimental paradigms, the absence of external events limits the effectiveness of resolving such emotional task-specific prefrontal and brain-wide electroencephalographic dynamics. To address this issue, we analyzed 64-channel EEG data acquired from 55 healthy and 55 depressed participants during a three-polar (positive, neutral and negative) emotional auditory task in a prefrontal internal event-driven approach. Specifically, we extracted depression-related prefrontal internal events according to signals acquired from Fp1, Fpz, and Fp2 by statistically comparing the microstates identified with an unsupervised Gaussian mixture model. We then examined such event-related neural dynamics in terms of power change in neural oscillations and cross-oscillation interaction in a reconstructed brain source space. Our results suggested that the prefrontal events can be used to resolve abnormal neural dynamics in depression. Our results revealed emotional process-specific and internal event-related abnormalities dynamically in the brain source space, as evaluated by internal event-related changes in power of neural oscillations and in modulations of low frequency phases to high frequency amplitudes. Overall, our findings present new insights for dynamical electroencephalographic biomarkers in depression, which potentially provide EEG signal decoding solutions for EEG feedback-based closed loop intervention of depression. © 2024, The Authors. All rights reserved.

关键词： Electroencephalography

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：