检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Liu, Xiaolan Yu, Jiadong Liu, Yuanwei Gao, Yue Mahmoodi, Toktam Lambotharan, Sangarapillai Tsang, Danny H.K. The Institute for Digital Technologies Loughborough University LondonE20 3BS United Kingdom Guangdong Guangzhou511400 China The School of Electronic Engineering and Computer Science Queen Mary University of London LondonE1 4NS United Kingdom The School of Computer Science Fudan University Shanghai200438 China The Department of Engineering King's College London LondonWC2R 2LS United Kingdom The Wolfson School of Mechanical Electrical and Manufacturing Engineering Loughborough University LoughboroughLE11 3TU United Kingdom The Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Clear Water Bay Hong Kong

The cloud-based solutions are becoming inefficient due to considerably large time delays, high power consumption, security and privacy concerns caused by billions of connected wireless devices and typically zillions bytes of data they produce at the network edge. A blend of edge computing and Artificial Intelligence (AI) techniques could optimally shift the resourceful computation servers closer to the network edge, which provides the support for advanced AI applications (e.g., video/audio surveillance and personal recommendation system) by enabling intelligent decision making on computing at the point of data generation as and when it is needed, and distributed Machine Learning (ML) with its potential to avoid the transmission of large dataset and possible compromise of privacy that may exist in cloud-based centralized learning. Therefore, AI is envisioned to become native and ubiquitous in future communication and networking systems. In this paper, we conduct a comprehensive overview of recent advances in distributed intelligence in wireless networks under the umbrella of native-AI wireless networks, with a focus on the basic concepts of native-AI wireless networks, on the AI-enabled edge computing, on the design of distributed learning architectures for heterogeneous networks, on the communication-efficient technologies to support distributed learning, and on the AI-empowered end-to-end communications. We highlight the advantages of hybrid distributed learning architectures compared to the state-of-art distributed learning techniques. We summarize the challenges of existing research contributions in distributed intelligence in wireless networks and identify the potential future opportunities. Copyright © 2022, The Authors. All rights reserved.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Reliability Benchmarking Through Energy System Optimization, Reliability and Seasonal Energy Storage Triangulation for Net Zero Electricity

SSRN

引用

SSRN 2023年

作者： Sadhukhan, Jhuma Sen, Sohum Randriamahefasoa, T.M.S. Centre for Environment & Sustainability School of Sustainability Civil and Environmental Engineering University of Surrey Guildford SurreyGU2 7XH United Kingdom Computer Science Department University College London Gower Street LondonWC1E 6BT United Kingdom

A regional/local energy planning model for optimum sustainable net-zero electricity (NZE) mix under spatiotemporal climate/meteorological and electrical load demand constraints comprises non-linear constrained mathematical programming (NLP), reliability against stored energy discharge analysis (RADA) and seasonal energy storage (SES) analyses. Reliability is the ratio between excess stored renewable power discharge and unmet load demand, a measure of the extent of unmet load demand met by excess stored renewable power. Coupled with the NLP, RADA and energy storage evaluations determine the SES conditions and realistic renewable proportions for NZE. Direct wind and solar can meet 63%, 62%, and 55% (15.8, 10.5 and 20.9 TWh/year) of electricity demand in South West, Greater London, and South East, UK, in 2035. Further, battery energy storage systems (BESS) increased 5.3, 3.7 and 4.9 TWh/year of wind and solar in the NZE mix in the three regions. The unmet demands (4, 2.7 and 12.2 TWh/year) in the three regions can be met by renewables with SES, compressed air energy storage (CAES) with upto 2 years of storage and pumped hydro, followed by gas with CCUS, bioenergy and renewables with hydrogen SES, respectively, in the increasing order of costs per lifecycle climate impact potential. The least cost NZE mix is wind and solar, 52 and 28 TWh/year, with BESS and CAES (13.9 and 18.9 TWh/year) at 91 EURO/MWh in the UK-South in 2035. However, infinitesimal reliability above 96 hours of energy storage can make gas with CCUS or bioenergy more attractive than renewables with SES. © 2023, The Authors. All rights reserved.

关键词： Life cycle

来源：评论

学校读者我要写书评

暂无评论

A wearable Gait Assessment Method for Lumbar Disc Herniation Based on Adaptive Kalman Filtering

arXiv

引用

arXiv 2023年

作者： Wang, Yongsong Li, Zhixin Guo, Zhaohui Ding, Yin Huan, Zhan Chen, Lin School of Computer Science and Artificial Intelligence Changzhou University Jiangsu Changzhou213000 China School of Microelectronics and Control Engineering Changzhou University Jiangsu Changzhou213000 China Department of Orthopedics The Affiliated Changzhou No.2 People’s Hospital of Nanjing Medical University 29 Xinglong Alley Changzhou China

Lumbar disc herniation (LDH) is a prevalent orthopedic condition in clinical practice. Inertial measurement unit sensors (IMUs) are an effective tool for monitoring and assessing gait impairment in patients with lumbar disc herniation (LDH). However, the current gait assessment of LDH focuses solely on single-source acceleration signal data, without considering the diversity of sensor data. It also overlooks the individual differences in motor function deterioration between the healthy and affected lower limbs in patients with LDH. To address this issue, we developed an LDH gait feature model that relies on multi-source adaptive Kalman data fusion of acceleration and angular velocity. We utilized an adaptive Kalman data fusion algorithm for acceleration and angular velocity to estimate the attitude angle and segment the gait phase. Two Inertial Measurement Units (IMUs) were used to analyze the gait characteristics of patients with lumbar disc issues and healthy individuals. This analysis included 12 gait characteristics, such as gait spatiotemporal parameters, kinematic parameters, and expansibility index numbers. Statistical methods were employed to analyze the characteristic model and confirm the biological differences between the healthy affected side of LDH and healthy subjects. Finally, a classifier based on feature engineering was utilized to classify the gait patterns of the affected side of patients with lumbar disc disease and healthy subjects. This approach achieved a classification accuracy of 95.50%, enhancing the recognition of LDH and healthy gait patterns. It also provided effective gait feature sets and methods for assessing LDH clinically. Copyright © 2023, The Authors. All rights reserved.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

3d Point Cloud Representation Learning Via Dual Curve-Graph Feature Fusion

SSRN

引用

SSRN 2024年

作者： Putra, Oddy Virgantara Priyadi, Ardyono Ogata, Kohichi Yuniarno, Eko Mulyanto Purnomo, Mauridhi Hery Department of Electrical Engineering Institut Teknologi Sepuluh Nopember Teknik Kimia Street East Java Surabaya60111 Indonesia Department of Informatics Universitas Darussalam Gontor Jalan Raya Siman East Java Ponorogo63471 Indonesia Faculty of Advanced Science and Technology Kumamoto University 2 Chome Chuo-ku Kurokami Kumamoto-ken Kumamoto860-0862 Japan Department of Computer Engineering Institut Teknologi Sepuluh Nopember Teknik Kimia Street East Java Surabaya60111 Indonesia

Point cloud data in three dimensions is extensively utilized in various domains like computer vision, augmented reality, and construction. Despite its accuracy and comprehensive detail, the unstructured nature of this data presents challenges for analysis. To tackle these issue, we introduces a novel method incorporates graph-based and curve-based representations to enhance representation learning for 3D point clouds. The proposed method integrates multiple specialized modules, including Curve Grouping , Curve Aggregation, Local Point-Feature Aggregation , and Geometry Disentanglement Attention, into a unified framework. This dual approach allows the model to capture local and global geometric features effectively. The model transforms point cloud data into graphs and curves, distilling sharp and gentle features, and then aggregating these features for advanced tasks like classification and segmentation. The efficacy of the proposed model was demonstrated through experiments on benchmark datasets for point cloud classification and segmentation. In point cloud classification using the ModelNet40 dataset, the DCGRL model achieved the highest accuracy of 92.4%, outperforming other well-known models. Similarly, in part segmentation, the DCGRL model obtained the highest mean Intersection over Union (mIoU) score of 86, indicating superior performance compared to other state-of-the-art methods. These results highlight the significant improvements the DCGRL approach offers in capturing intricate geometric details and enhancing the overall accuracy and robustness of point cloud processing tasks. © 2024, The Authors. All rights reserved.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Semantic Communications for Future Internet: Fundamentals, Applications, and Challenges

arXiv

引用

arXiv 2022年

作者： Yang, Wanting Du, Hongyang Liew, Zi Qin Lim, Wei Yang Bryan Xiong, Zehui Niyato, Dusit Chi, Xuefen Shen, Xuemin Sherman Miao, Chunyan The Department of Communications Engineering Jilin University Changchun130012 China Information Systems Technology and Design Pillar Singapore University of Technology and Design Singapore Singapore The School of Computer Science and Engineering The Energy Research Institute Ntu Interdisciplinary Nanyang Technological University Singapore Singapore Alibaba Group The Alibaba-NTU Joint Research Institute Nanyang Technological University Singapore Singapore School of Computer Science and Engineering Ntu Singapore Singapore The Department of Electrical and Computer Engineering University of Waterloo WaterlooONN2L 3G1 Canada The School of Computer Science and Engineering Nanyang Technological University Singapore Singapore The Alibaba-NTU Joint Research Institute Nanyang Technological University Singapore Singapore The Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly Nanyang Technological University Singapore Singapore

With the increasing demand for intelligent services, the sixth-generation (6G) wireless networks will shift from a traditional architecture that focuses solely on a high transmission rate to a new architecture that is based on the intelligent connection of everything. Semantic communication (SemCom), a revolutionary architecture that integrates user as well as application requirements and the meaning of information into data processing and transmission, is predicted to become a new core paradigm in 6G. While SemCom is expected to progress beyond the classical Shannon paradigm, several obstacles need to be overcome on the way to a SemCom-enabled smart wireless Internet. In this paper, we first highlight the motivations and compelling reasons of SemCom in 6G. Then, we provide an overview of SemCom-related theory development. After that, we introduce three types of SemCom, i.e., semantic-oriented communication, goal-oriented communication, and semantic-aware communication. Following that, we organize the design of the communication system into three dimensions, i.e., semantic information (SI) extraction, SI transmission, and SI metrics. For each dimension, we review existing techniques and discuss their benefits and limitation, as well as the remaining challenges. Then, we introduce the potential applications of SemCom in 6G and portray the vision of future SemCom-empowered network architecture. Finally, we outline future research opportunities. In a nutshell, this paper provides a holistic review of the fundamentals of SemCom, its applications in 6G networks, and the existing challenges and open issues with insights for further in-depth investigations. Copyright © 2022, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Reversible Graph Neural Network-based Reaction Distribution Learning for Multiple Appropriate Facial Reactions Generation

arXiv

引用

arXiv 2023年

作者： Xu, Tong Spitale, Micol Tang, Hao Liu, Lu Gunes, Hatice Song, Siyang The School of Computing and Mathematical Sciences University of Leicester LeicesterLE2 7RH United Kingdom The AFAR Lab Department of Computer Science and Technology University of Cambridge CambridgeCB3 0FT United Kingdom The Department of Information Technology and Electrical Engineering ETH Zurich Zurich8092 Switzerland

Generating facial reactions in a human-human dyadic interaction is complex and highly dependent on the context since more than one facial reactions can be appropriate for the speaker's behaviour. This has challenged existing machine learning (ML) methods, whose training strategies enforce models to reproduce a specific (not multiple) facial reaction from each input speaker behaviour. This paper proposes the first multiple appropriate facial reaction generation framework that re-formulates the one-to-many mapping facial reaction generation problem as a one-to-one mapping problem. This means that we approach this problem by considering generating a distribution of listeners' appropriate facial reactions instead of multiple different appropriate facial reactions, i.e.,'many' appropriate facial reaction labels are summarised as'one' distribution label during training. Our model consists of a perceptual processor, a cognitive processor, and a motor processor. The motor processor is implemented with a novel Reversible Multi-dimensional Edge Graph Neural Network (REGNN). This allows us to obtain a distribution of appropriate real facial reactions during the training process, enabling the cognitive processor to be trained to predict the appropriate facial reaction distribution. At the inference stage, the REGNN decodes an appropriate facial reaction by using the predicted distribution as input. Experimental results demonstrate that our approach outperforms existing models in generating more appropriate, realistic, and synchronized facial reactions. The improved performance is largely attributed to the proposed appropriate facial reaction distribution learning strategy and the use of a REGNN. The code will be made publicly available at https://***/TongXu-05/*** Codes 68T40 © 2023, CC0.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Covid-19 detection using medical images

引用

AIP Conference Proceedings 2023年第1期2591卷

作者： Sabah Abdulazeez Jebur Mohanad Ali Mohammed Alia Karim Abdulhassan 1Department of Computer Techniques Engineering Imam Al-Kadhum College Baghdad Iraq 2Computer Science Department University of Technology Baghdad Iraq

At the present time, COVID-19 considers the pandemic that has most dangerous, representing a great threat to both, Communities and individuals especially scientist and researchers. Determine the infection in the early time is the only way to avoid getting worst especially for those who have respiratory diseases. The researchers working daily without stopping to get the possible solutions to rescue patients or who have the Symptoms of the disease. CT-Scans and X-rays for the lungs consider one of the most popular methods used by the researchers, where analyzing this image to detect the infection by this pandemic. However, it demands long time to examine each test, beside a large number of radiology specialists will be *** data may be an interesting signal data samples that can be used to detect the COVID-19 via number of oxygen blood tests, in this research, detection of COVID-19 through some tests of oxygen signals is an essential goal. Tests like oxygen saturation (spO2), Pulse Oximetry (plusOX), Arterial blood gas (ABG) and Carboxyhaemoglobin (COHb) are the most common and important test for this purpose. In this research, a method based on deep neural network as well as machine learning methods using (stochastic gradient decision, decision tree, random forest, linear regression and k-nearest neighbour) was proposed for detecting COVID-19 through analysing the X-rays of patients which will look for visual indicators found in the chest radiography imaging of COVID-19 patients. Also proposed architecture can be applied to the oxygen data when it will be available (lack of well-structured data).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Densenet Scheduling for Heterogeneous Parallel Systems

SSRN

引用

SSRN 2022年

作者： Liu, Pangfeng Cheng, Yu-Che Wu, Jan-Jan Department of Computer Science and Information Engineering National Taiwan University No. 1 Sec. 4 Roosevelt Rd. Taipei10617 Taiwan Institute of Information Science Academia Sinica 128 Academia Road Sec. 2 Nangang11529 Taiwan

This paper discusses the scheduling problem for DenseNet on a heterogeneous *** is a promising network topology for AI *** has extensive data dependency, so it is challenging to schedule the tasks among processing ***, modern computing systems have heterogeneous processors for different computing needs, and communication among processors is ***, it is essential to schedule the computation tasks to the suitable processors while reducing the communication among *** show that the DenseNet scheduling problem is NP-Complete when the number of heterogeneous processors is $n^2$, where $n$ is the size of the ***, we show that when there are only {\em two} heterogeneous processors, dynamic programming can schedule the DenseNet computation optimally in $O(n)$ time by a series of *** heterogeneous system model is consistent with the current computer systems, with one CPU and one GPU as the primary resources for AI *** also implement these algorithms and compare their timing *** experiment results indicate that the execution time of these algorithms concurs with the theoretical analysis. The dynamic programming relies on two keen observations to improve the time ***, we observe that the delay effect due to communication will not affect later tasks assigned to the same processor. Second, we observe that we can simplify the dependency by considering one task in increasing time order and performing an amortized cost analysis for all tasks. Finally, we derive a pseudo-polynomial-time algorithm when the number of heterogeneous processors is a small bounded constant $R$.The algorithm runs in $O(n D^{R-1})$, where $D$ is the sum of all communication and computation *** algorithm is practical since the current resource includes few possibilities, e.g., CPU, GPU, FPGA, and Tensor cores. © 2022, The Authors. All rights reserved.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Efficient Learning of Continuous-Variable Quantum States

arXiv

引用

arXiv 2023年

作者： Wu, Ya-Dong Zhu, Yan Chiribella, Giulio Liu, Nana John Hopcroft Center for Computer Science Shanghai Jiao Tong University Shanghai200240 China QICI Quantum Information and Computation Initiative Department of Computer Science The University of Hong Kong Pokfulam Road Hong Kong Department of Computer Science Parks Road OxfordOX1 3QD United Kingdom Perimeter Institute for Theoretical Physics WaterlooONN2L 2Y5 Canada Institute of Natural Sciences Shanghai Jiao Tong University Shanghai200240 China Ministry of Education Key Laboratory in Scientific and Engineering Computing Shanghai Jiao Tong University Shanghai200240 China School of Mathematical Sciences Shanghai Jiao Tong University Shanghai200240 China University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai200240 China

The characterization of continuous-variable quantum states is crucial for applications in quantum communication, sensing, simulation and computing. However, a full characterization of multimode quantum states requires a number of experiments that grows exponentially with the number of modes. Here we propose an alternative approach where the goal is not to reconstruct the full quantum state, but rather to estimate its characteristic function at a given set of points. For multimode states with reflection symmetry, we show that the characteristic function at M points can be estimated using only O(log M) copies of the state, independently of the number of modes. When the characteristic function is known to be positive, as in the case of squeezed vacuum states, the estimation is achieved by an experimentally friendly setup using only beamsplitters and homodyne measurements. © 2023, CC BY-NC-SA.

关键词： Quantum communication

来源：评论

学校读者我要写书评

暂无评论

Sparse LiDAR and stereo fusion (SLS-Fusion) for depth estimation and 3D object detection 11

Sparse LiDAR and stereo fusion (SLS-Fusion) for depth estima...

引用

11th International Conference of Pattern Recognition Systems, ICPRS 2021

作者： Mai, Nguyen-Anh-Minh Duthon, Pierre Khoudour, Louahdi Crouzil, Alain Velastin, Sergio A. Cerema Equipe-projet STI 1 Avenue du Colonel Roche Toulouse31400 France Institut de Recherche en Informatique de Toulouse Université de Toulouse UPS Toulouse31062 France Cerema Equipe-projet STI 8-10 rue Bernard Palissy Clermont-Ferrand Cedex 263017 France Department of Computer Science and Engineering Universidad Carlos III de Madrid Madrid Leganés28911 Spain School of Electronic Engineering and Computer Science Queen Mary University of London United Kingdom

ISBN: (纸本)9781839534300

The ability to accurately detect and localize objects is recognized as being the most important for the perception of self-driving cars. From 2D to 3D object detection, the most difficult is to determine the distance from the ego-vehicle to objects. Expensive technology like LiDAR can provide a precise and accurate depth information, so most studies have tended to focus on this sensor showing a performance gap between LiDAR-based methods and camera-based methods. Although many authors have investigated how to fuse LiDAR with RGB cameras, as far as we know there are no studies to fuse LiDAR and stereo in a deep neural network for the 3D object detection task. This paper presents SLS-Fusion, a new approach to fuse data from 4-beam LiDAR and a stereo camera via a neural network for depth estimation to achieve better dense depth maps and thereby improves 3D object detection performance. Since 4-beam LiDAR is cheaper than the well-known 64-beam LiDAR, this approach is also classified as a low-cost sensors-based method. Through evaluation on the KITTI benchmark, it is shown that the proposed method significantly improves depth estimation performance compared to a baseline method. Also when applying it to 3D object detection, a new state of the art on low-cost sensor based method is achieved. © 2021 Institution of engineering and Technology. All rights reserved.

关键词： Autonomous vehicles

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：