检索结果-内蒙古大学图书馆

CoLLaRS : A cloud–edge–terminal collaborative lifelong learning framework for AIoT

Future Generation computer systems 2024年 158卷 447-456页

作者： Hu, Shijing Lin, Junxiong Lu, Zhihui Du, Xin Duan, Qiang Huang, Shih-Chia School of Computer Science Fudan University Shanghai200438 China Information Sciences and Technology Department Pennsylvania State University AbingtonPA19001 United States Department of Electronic Engineering National Taipei University of Technology Taipei10608 Taiwan

AIoT applications often encounter challenges such as terminal resource constraints, data drift, and data heterogeneity in real world, leading to problems such as catastrophic forgetting, low generalization ability, and low accuracy during model training. To address these challenges, we proposed CoLLaRS, a cloud–edge–terminal collaborative lifelong learning framework for AIoT applications. In the CoLLaRS framework, we alleviate the problem of terminal resource constraints by uploading terminal tasks at the edge. CoLLaRS uses continuous training at the edge to achieve lifelong learning training of the model and solve the problem of catastrophic forgetting. CoLLaRS employs federated optimization in the cloud to perform personalized aggregation of different edge models and solve the problem of weak model generalization ability. Finally, the model is fine-tuned at the terminal to further optimize its accuracy in local tasks. Our experiments on real-world datasets showed that CoLLaRS has an 8% improvement in accuracy and a 5% improvement in backward transfer(BWT) and forward transfer(FWT) compared to other baseline algorithms. The results of the ablation experiments further confirmed the effectiveness of CoLLaRS. © 2024 Elsevier B.V.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Control and Stability of Large-scale Power System with Highly Distributed Renewable Energy Generation:Viewpoints from Six Aspects

引用

CSEE Journal of Power and Energy systems 2023年第1期9卷 8-14页

作者： Qing-Hua Wu Anjan Bose Chanan Singh Joe H.Chow Gang Mu Yuanzhang Sun Zhaoxi Liu Zhigang Li Yang Liu School of Electric Power Engineering South China University of TechnologyGuangzhou 510640China School of Electrical Engineering&Computer Science Washington State UniversityPullmanWA 99164USA Department of Electrical and Computer Engineering Texas A&M UniversityCollege StationTX 77843USA Department of Electrical Computer and Systems EngineeringRensselaer Polytechnic Institute(RPI)TroyNYUSA Key Laboratory of Modern Power System Simulation and Control&Renewable Energy Technology Ministry of Education(Northeast Electric Power University)Jilin 132012China School of Electrical Engineering and Automation Wuhan UniversityWuhan 430072China

Power systems are moving toward a low-carbon or carbon-neutral future where high penetration of renewables is *** conventional fossil-fueled synchronous generators in the transmission network being replaced by renewable energy generation which is highly distributed across the entire grid,new challenges are emerging to the control and stability of large-scale power *** analysis and control methods are needed for power systems to cope with the ongoing *** the CSEE JPES forum,six leading experts were invited to deliver keynote speeches,and the participating researchers and professionals had extensive exchanges and discussions on the control and stability of power ***,potential changes and challenges of power systems with high penetration of renewable energy generation were introduced and explained,and advanced control methods were proposed and analyzed for the transient stability enhancement of power grids.

关键词： Distributed energy resources high converter penetration power system control renewable energy generation stability control transient stability

来源：评论

学校读者我要写书评

暂无评论

Oil Spill Detection in SAR Images based on Improved YOLOX-S

Oil Spill Detection in SAR Images based on Improved YOLOX-S

引用

2022 International Conference on computer engineering and Artificial Intelligence, ICCEAI 2022

作者： Xu, Fang-Yuan An, Xiang-Ze Liu, Wen-Qi School of Computer and Information Engineering Heilongjiang University of Science and Technology Harbin China

ISBN: (数字)9781665468039

ISBN: (纸本)9781665468039

In order to identify the oil spills rapidly and accurately in SAR images, a detection method of oil spills in SAR images based on improved YOLOX-S was proposed. First, data enhancement and other pretreatment measures were done for collected SAR images of oil spills;second, improved FFDNet was used to reduce the speckle noise in SAR images, thus reducing the influence of things which were similar to oil spills on experimental results;finally, in order to fully extract the features of oil spills in SAR images, ECA module was added into YOLOX-S model to improve the identification accuracy of oil spills in SAR images. The experimental results showed that the improved YOLOX-S model could reduce the speckle noise in SAR images to some extent and identify quickly and accurately the oil spills in SAR images with accuracy rate of 86.76%. © 2022 IEEE.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

McCrolin: Multi-consistency Cross-lingual Training for Retrieval Question Answering

McCrolin: Multi-consistency Cross-lingual Training for Retri...

引用

2024 Findings of the Association for Computational Linguistics, EMNLP 2024

作者： Limkonchotiwat, Peerat Ponwitayarat, Wuttikorn Lowphansirikul, Lalita Manakul, Potsawee Udomcharoenchaikit, Can Chuangsuwanich, Ekapol Nutanong, Sarana AI Singapore Singapore SCB 10X Thailand School of Information Science and Technology VISTEC Thailand Department of Computer Engineering Chulalongkorn University Thailand

ISBN: (纸本)9798891761681

Automated question answering (QA) systems are increasingly relying on robust cross-lingual retrieval to identify and utilize information from multilingual sources, ensuring comprehensive and contextually accurate responses. Existing approaches often struggle with consistency across multiple languages and multi-size input scenarios. To address these challenges, we propose McCrolin, a Multi-consistency Cross-lingual training framework, leveraging multi-task learning to enhance cross-lingual consistency, ranking stability, and input-size robustness. Experimental results demonstrate that McCrolin achieves state-of-the-art performance on standard cross-lingual retrieval QA datasets. Furthermore, McCrolin outperforms competitors when dealing with various input sizes on downstream tasks. In terms of generalizability, results from further analysis show that our method is effective for various encoder architectures and sizes. Codes and models are available at https://***/mrpeerat/McCrolin. © 2024 Association for Computational Linguistics.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

A graph attention feature pyramid network for 3D object detection in point clouds 7

A graph attention feature pyramid network for 3D object dete...

引用

7th International Conference on Intelligent Informatics and Biomedical sciences, ICIIBMS 2022

作者： Wenju, Li Wanghui, Chu Liu, Cui Gan, Zhang Shanghai Institute of Technology School of Computer Science and Information Engineering Shanghai China

ISBN: (数字)9781665482295

ISBN: (纸本)9781665482295

Graph Neural Networks have been recently applied to 3D object detection in point clouds. The works, however, have the problem of insufficient detection accuracy for small objects and objects in complex backgrounds. Towards this end, a graph attention feature pyramid network is proposed for 3D point clouds object detection. Specifically, the network constructs a near-neighbors graph in a point cloud which is downsampled;and then, a graph attention feature pyramid network is designed to extract features of the point cloud at different levels;finally, a feature fusion module is employed to fuse the features before point classification and object detection. Compared with the benchmark network Point-GNN on the KITTI dataset, the detection accuracy of cars in complex scenes is improved by 2.53%, and the detection accuracy of pedestrian and cyclist categories in moderate scenes and complex scenes is improved by 5.17%. and 3.62%. The experiments show that the designed method is more effective for the detection of small objects and objects in complex backgrounds in 3D object detection. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A Coarse to Fine Detection Method for Prohibited Object in X-ray Images Based on Progressive Transformer Decoder 24

A Coarse to Fine Detection Method for Prohibited Object in X...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Ma, Chunjie Du, Lina Gao, Zan Zhuo, Li Wang, Meng Shandong Jinan China School of Computer Science and Technology Shandong Jianzhu University Shandong Jinan China Faculty of Information Technology Beijing University Of Technology Beijing China School of Computer Science and Information Engineering Hefei University of Technology Anhui Hefei China Key Laboratory of Computer Vision and System Ministry of Education Tianjin University of Technology Tianjin300384 China

ISBN: (纸本)9798400706868

Currently, Transformer-based prohibited object detection methods in X-ray images appear constantly, but there are still some shortcomings such as poor performance and high computational complexity for prohibited object detection with heavily occlusion. Therefore, a coarse to fine detection method for prohibited object in X-ray images based on progressive Transformer decoder is proposed in this paper. Firstly, a coarse to fine framework is proposed, which includes two stages: coarse detection and fine detection. Through adaptive inference in stages, the computational efficiency of the model is effectively improved. Then, a position and class object queries method is proposed, which improves the convergence speed and detection accuracy of the model by fusing the position and class information of prohibited object with object queries. Finally, a progressive Transformer decoder is proposed, which distinguishes high and low score queries by decreasing confidence thresholds, so that high-score queries are not affected by low-score queries in the decoding stage, and the model can focus more on decoding low-score queries, which usually correspond to prohibited object with severe occlusion. The experimental results on three public benchmark datasets (SIXray, OPIXray, HiXray) demonstrate that compared with the baseline DETR, the proposed method achieves the state-of-the-art detection accuracy with a 21.6% reduction in model computational complexity. Especially for prohibited objects with heavily occlusion, accurate detection can be carried out. © 2024 ACM.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

An Attention Mechanism Based AVOD Network for 3D Vehicle Detection

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2023年 1-13页

作者： Yuan, Di Zhang, Haiping Shu, Xiu Liu, Qiao Chang, Xiaojun He, Zhenyu Shi, Guangming Guangzhou Institute of Technology Xidian University Xian China School of Computer Science and Cyber Engineering Guangzhou University Guangzhou China National Center for Applied Mathematics Chongqing Normal University Chongqing China Faculty of Engineering and Information Technology University of Technology Sydney Broadway New South Wales Australia School of Computer Science and Technology Harbin Institute of Technology Shenzhen China

With the continuous advancement of autonomous driving technology, 3D vehicle detection has become of widespread interest. The traditional aggregate view object detection (AVOD) framework has achieved some good results in 3D vehicle detection tasks. However, the complexity of the 3D vehicle detection scenario makes the current detection methods still not meet the actual requirements. To enhance the detection accuracy of 3D vehicle targets, we propose to equip an attention mechanism to improve the representation capability of feature maps, thereby further increasing the precision of 3D vehicle detection. Specifically, we have added the channel attention ECANet, spatial attention SANet, and mixed attention ECANet+SANet respectively into the image-based feature pyramid network of the AVOD detection framework, which can enhance the feature maps representation and improve the detection accuracy observably. The improved AVOD network is verified using the KITTI dataset. By showing the detection results of these attention mechanisms, it is found that the feature pyramid networks (FPN) module in the AVOD network based on Image has the best performance when integrating a mixed attention mechanism. In comparison to the original AVOD network, the detection results on the average precision index of the proposed method have improved by 2.29%, 2.81%, and 1.32% in the three indexes of simple, medium, and difficult, respectively. Extensive experiments have confirmed the practicality and efficacy of the AVOD network to equip the attention mechanisms for 3D vehicle detection. IEEE

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Task Offloading Optimization for AGVs with Fixed Routes in Industrial IoT Environment

引用

China Communications 2023年第5期20卷 302-314页

作者： Peng Liu Zifu Wu Hangguan Shan Fei Lin Qi Wang Qingshan Wang School of Computer Science and Technology Hangzhou Dianzi UniversityHangzhou 310018China HDU-ITMO Joint Institute Hangzhou Dianzi UniversityHangzhou 310018China College of Information Science and Electronic Engineering Zhejiang UniversityHangzhou 310027China Hefei University of Technology School of MathematicsHefei 230009China

In order to solve the delay requirements of computing intensive tasks in industrial Internet of things,edge computing is moving from theoretical research to practical *** servers(ESs)have been deployed in factories,and on-site auto guided vehicles(AGVs),besides doing their regular transportation tasks,can partly act as mobile collectors and distributors of computing data and *** AGVs may offload tasks to the same ES if they have overlapping path segments,resource allocation conflicts are *** this paper,we study the problem of efficient task offloading from AGVs to ESs,along their fixed *** propose a multi-AGV task offloading optimization algorithm(MATO),which first uses the weighted polling algorithm to preliminarily allocate tasks for individual AGVs based on load balancing,and then uses the Deep Q-Network(DQN)model to obtain the updated offloading strategy for the AGV *** simulation results show that,compared with the existing methods,the proposed MATO algorithm can significantly reduce the maximum completion time of tasks and be stable under various parameter settings.

关键词： industrial Internet of Things task offloading optimization auto guided vehicles reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Collaborative filtering recommendation algorithm combining tag information and rating difference 6

Collaborative filtering recommendation algorithm combining t...

引用

6th IEEE information Technology and Mechatronics engineering Conference, ITOEC 2022

作者： Ji, Lu Yu, Wanjun Chen, Ying School of Computer Science and Information Engineering Shanghai Institute of Technology Shanghai China

ISBN: (纸本)9781665431859

Aiming at the problem of the low calculation accuracy of similarity between users in the traditional collaborative filtering recommendation algorithm, a collaborative filtering recommendation algorithm combining tag information and rating difference is proposed. Firstly, the algorithm introduces time weights to describe the dynamic change of users' interest, and defines users' preference values for tags according to tag information so as to calculate the interest similarity between users. Secondly, considering the influence of rating difference between users on recommendation effect, the algorithm uses the mean ratings as the benchmarks to calculate the rating difference between them. Finally, the algorithm comprehensively considers interest similarity and rating difference between users, generates target user's nearest neighbors and makes recommendations. The experimental results based on movieLens dataset show that the proposed algorithm effectively improves the calculation accuracy of similarity between users and the recommendation performance. © 2022 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

DP-BEGAN: A Generative Model of Differential Privacy Algorithm

DP-BEGAN: A Generative Model of Differential Privacy Algorit...

引用

2022 International Conference on computer engineering and Artificial Intelligence, ICCEAI 2022

作者： Shi, Er-Mei Liu, Jia-Xi Ji, Yuan-Ming Chang, Liang School of Computer and Information Engineering Heilongjiang University of Science and Technology Harbin China

ISBN: (数字)9781665468039

ISBN: (纸本)9781665468039

In recent years, differential privacy has gradually become a standard definition in the field of data privacy protection. Differential privacy does not need to make assumptions about the prior knowledge of privacy adversaries, so it has a more stringent effect than existing privacy protection models and definitions. This good feature has been used by researchers to solve the in-depth learning problem restricted by the problem of privacy and security, making an important breakthrough, and promoting its further large-scale application. Combining differential privacy with BEGAN, we propose the DP-BEGAN framework. The differential privacy is realized by adding carefully designed noise to the gradient of Gan model training, so as to ensure that Gan can generate unlimited synthetic data that conforms to the statistical characteristics of source data and does not disclose privacy. At the same time, it is compared with the existing methods on public datasets. The results show that under a certain privacy budget, this method can generate higher quality privacy protection data more efficiently, which can be used in a variety of data analysis tasks. The privacy loss is independent of the amount of synthetic data, so it can be applied to large datasets. © 2022 IEEE.

关键词： Data privacy

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：