检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zhang, Chen Li, Guorong Qi, Yuankai Ye, Hanhua Qing, Laiyun Yang, Ming-Hsuan Huang, Qingming State Key Laboratory of Information Security Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China School of Computer Science and Technology University of Chinese Academy of Sciences China Australian Institute for Machine Learning The University of Adelaide Australia University of California Merced United States Key Laboratory of Intelligent Information Processing Institute of Computing Technology CAS China

The goal of weakly supervised video anomaly detection is to learn a detection model using only video-level labeled data. However, prior studies typically divide videos into fixed-length segments without considering the complexity or duration of anomalies. Moreover, these studies usually just detect the most abnormal segments, potentially overlooking the completeness of anomalies. To address these limitations, we propose a Dynamic Erasing Network (DE-Net) for weakly supervised video anomaly detection, which learns multi-scale temporal features. Specifically, to handle duration variations of abnormal events, we first propose a multi-scale temporal modeling module, capable of extracting features from segments of varying lengths and capturing both local and global visual information across different temporal scales. Then, we design a dynamic erasing strategy, which dynamically assesses the completeness of the detected anomalies and erases prominent abnormal segments in order to encourage the model to discover gentle abnormal segments in a video. The proposed method obtains favorable performance compared to several state-of-the-art approaches on three datasets: XD-Violence, TAD, and UCF-Crime. Code will be made available at https://***/ArielZc/DE-Net. Copyright © 2023, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Distributed Quantum Simulation

arXiv

引用

arXiv 2024年

作者： Feng, Tianfeng Xu, Jue Yu, Wenjun Ye, Zekun Yao, Penghui Zhao, Qi QICI Quantum Information and Computation Initiative Department of Computer Science School of Computing and Data Science The University of Hong Kong Pokfulam Road Hong Kong State Key Laboratory for Novel Software Technology Nanjing University Hefei National Laboratory Nanjing 210023 China Hefei230088 China

Quantum simulation is a promising pathway toward practical quantum advantage by simulating large-scale quantum systems. In this work, we propose communication-efficient distributed quantum simulation protocols by exploring three quantum simulation algorithms, including the product formula, the truncated Taylor series, and the processing of quantum signals over a quantum network. Our protocols are further shown to be optimal by deriving a lower bound on the quantum communication complexity for distributed quantum simulations with respect to evolution time and the number of distributed quantum processing units. Additionally, our distributed techniques go beyond quantum simulation and are applied to distributed versions of Grover’s algorithms and quantum phase estimation. Our work not only paves the way for achieving a practical quantum advantage by scalable quantum simulation but also enlightens the design of more general distributed architectures across various physical systems for quantum computation. Copyright © 2024, The Authors. All rights reserved.

关键词： Quantum computers

来源：评论

学校读者我要写书评

暂无评论

OneForecast: A Universal Framework for Global and Regional Weather Forecasting

arXiv

引用

arXiv 2025年

作者： Gao, Yuan Wu, Hao Shu, Ruiqi Dong, Huanshuo Xu, Fan Chen, Rui Yan, Yibo Wen, Qingsong Hu, Xuming Wang, Kun Wu, Jiahao Li, Qing Xiong, Hui Huang, Xiaomeng Department of Earth System Science Ministry of Education Key Laboratory for Earth System Modeling Institute for Global Change Studies Tsinghua University China TEG Tencent China Department and Computer and Science University of Science and Technology of China China Institute for Interdisciplinary Information Sciences Tsinghua University China Department of Computer Science and Engineering The HongKong University of Science and Technology Hong Kong AI Thrust The Hong Kong University of Science and Technology Guangzhou China Squirrel Ai Learning China School of Computer Science and Engineering Nanyang Technological University Singapore Department of Computing The Hong Kong Polytechnic University Hong Kong

Accurate weather forecasts are important for disaster prevention, agricultural planning, and water resource management. Traditional numerical weather prediction (NWP) methods offer physically interpretable high-accuracy predictions but are computationally expensive and fail to fully leverage rapidly growing historical data. In recent years, deep learning methods have made significant progress in weather forecasting, but challenges remain, such as balancing global and regional high-resolution forecasts, excessive smoothing in extreme event predictions, and insufficient dynamic system modeling. To address these issues, this paper proposes a global-regional nested weather forecasting framework based on graph neural networks (GNNs). By combining a dynamic system perspective with multi-grid theory, we construct a multi-scale graph structure and densify the target region to capture local high-frequency features. We introduce an adaptive information propagation mechanism, using dynamic gating units to deeply integrate node and edge features for more accurate extreme event forecasting. For high-resolution regional forecasts, we propose a neural nested grid method to mitigate boundary information loss. Experimental results show that the proposed method performs excellently across global to regional scales and short-term to long-term forecasts, especially in extreme event predictions (e.g., typhoons), significantly improving forecast accuracy. Our codes are available at https://***/YuanGao-YG/OneForecast. Copyright © 2025, The Authors. All rights reserved.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

Predictive ISAC Beamforming Using LoRA-Transformer LSTM Integrated with GNN in Ultra-Dense D2D mmWave Networks

引用

IEEE Transactions on Vehicular technology 2025年

作者： Liu, Haochen Bi, Jia Li, Bohan Xiang, Luping Yang, Xin Zhang, Zhaolin Wang, Ling Northwestern Polytechnical University School of Electronics and Information Xi'an710019 China Rutherford Appleton Laboratory Scientific Computing Department STFC DidcotOX11 0QX United Kingdom Ocean University of China Faculty of Information Science and Engineering Qingdao266100 China Nanjing University State Key Laboratory of Novel Software Technology Nanjing210008 China School of Intelligent Software and Engineering Suzhou China

Effective beamforming is crucial in ultra-dense device-to-device (D2D) mmWave networks to reduce interference and optimize communication. However, current methods often struggle to balance high data rates with minimal signaling overhead, especially with imperfect channel state information (CSI). This paper introduces a machine learning-assisted beamforming design tailored for integrated sensing and communication (ISAC), combining an efficient Low-Rank Adaptation Blocked LSTM Transformer with Message Passing Graph Neural Network that is called as LoRA_BLT-MPGNN. The LoRA_BLT component provides a key advantage by efficiently capturing both temporal and spatial patterns through fixed-weight updates during training and low-rank approximation for fine-tuning, significantly reducing computational complexity without sacrificing accuracy. Using historical channel data, the method implicitly captures channel characteristics and directly predicts the beamforming matrix for the subsequent time slot, minimizing signaling overhead while maximizing the achievable sum rate. Our research and simulation results show that the proposed predictive beamforming design maintains robust sensing performance and achieves a sum rate closely approaching the upper bound achievable with perfect CSI. Additionally, the incorporation of MPGNNs enables efficient modeling and learning of complex inter-node relationships, making the approach adaptable to networks of varying sizes and densities. Simulations further demonstrate that the LoRA_BLT-MPGNN model can be generalized across different network densities and coverage areas, reducing computational complexity by approximately 32× in terms of Tensor Floating Point Operations (TFLOPs) compared to conventional methods while maintaining high performance. © 2025 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning-based SDN Controller Framework for Drone Management

A Machine Learning-based SDN Controller Framework for Drone ...

引用

2021 IEEE Globecom Workshops, GC Wkshps 2021

作者： Yazdinejad, Abbas Rabieinejad, Elnaz Dehghantanha, Ali Parizi, Reza M. Srivastava, Gautam University of Guelph Cyber Science Lab School of Computer Science ON Canada K.N.Toosi University of Technology Information Technology Department Tehran Iran Kennesaw State University College of Computing and Software Engineering GA United States Brandon University Department of Mathematics and Computer Science MB Canada

ISBN: (纸本)9781665423908

With the advancement of information and communication technology, Unmanned Aerial Vehicles (UAV), popularly known as drones, have also increased. The drones have been noted for their wide range of applications such as military, search and rescue operation, disaster detection and monitoring, agriculture, and delivery. Each type of drone has different characteristics and functionality based on its application, making them a security threat for some city zone. Therefore, there is an essential need for efficient drone management based on their type and application in different zones. To do this, we proposed a Machine learning (ML) based Software Defined Network (SDN) drone management framework. In this framework, the SDN controller uses ML with the drone's radio frequency feature to detect its type and application and, according to its application, authenticate it and assign communication rules. SDN controller records authentication information in a DAG-based Distributed Ledger technology (DLT) available for other SDN controllers. When a drone desires to migrate to another zone, the destination SDN controller can achieve authentication information by referring to DAG-based DLT, and there is no need for re-authentication. The experimental result shows authentication delay reduction in our proposed framework. Moreover, we adopted ML algorithms includes Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Logistic Regression (LR), to evaluate our proposed framework in drone's type classification. The result shows that the RF algorithm shows the best performance with 92.81% accuracy in the classification of the drone's type. © 2021 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Hofstadter spectrum in a semiconductor moiré lattice

引用

Physical Review B 2025年第20期111卷 L201302-L201302页

作者： Chen Zhao Ming Wu Zhen Ma Miao Liang Ming Lu Jin-Hua Gao X. C. Xie School of Physics and Wuhan National High Magnetic Field Center Huazhong University of Science and Technology Wuhan Hubei 430074 China School of Electronics and Information Zhengzhou University of Light Industry Zhengzhou 450002 China Beijing Academy of Quantum Information Sciences Beijing 100193 China International Center for Quantum Materials School of Physics Peking University Beijing 100871 China Institute for Nanoelectronic Devices and Quantum Computing Fudan University Shanghai 200433 China Hefei National Laboratory Hefei 230088 China

Recently, the Hofstadter spectrum of a twisted WSe2/MoSe2 heterobilayer was observed in experiment [C. R. Kometter et al. Nat. Phys. 19, 1861 (2023)], but the origin of Hofstadter states remains unclear. Here, we present a comprehensive theoretical interpretation of the observed Hofstadter states by calculating their accurate Hofstadter spectrum. We point out that the valley Zeeman effect, a unique feature of the transition metal dichalcogenide (TMD) materials, plays a crucial role in determining the shape of the Hofstadter spectrum, due to the narrow bandwidth of the moiré bands. This is distinct from the graphene-based moiré systems. We further predict that the Hofstadter spectrum of the moiré flat band, which was not observed in the experiment, can be observed in the same system with a larger twist angle 2∘≲θ≲3∘. Our theory paves the way for further studies of the interplay between the Hofstadter states and correlated insulating states in such moiré lattice systems.

关键词： II-VI semiconductors

来源：评论

学校读者我要写书评

暂无评论

Digital climate formation amid the rise of artificial intelligence technology 24

Digital climate formation amid the rise of artificial intell...

引用

24th Pacific Asia Conference on information Systems: information Systems (IS) for the Future, PACIS 2020

作者： Sari, Rora P. Shi, Yunfei Pan, Shan L. Nguyen, Hoang D. School of Information Systems and Technology Management UNSW Sydney Australia School of Computing Science University of Glasgow Singapore

Work climate is a complex organisational phenomenon that has been studied from various perspectives in the past three decades. Despite its long tradition, research that pays specific attention to the work climate that exists in the digital economy is unjustifiably rare. Few studies have empirically investigated how a digital climate takes shape in an organisation. Against this backdrop, in this paper, we report on an interpretive case study that explores the formation of digital climate in an organisation developing artificial intelligence (AI) technology. The preliminary findings reveal learning and experimentation to be essential interactions as they allow conformity on the perception of work practices among digital workforces. Accordingly, our early findings show that this conformity enables a new type of work climate-which we have termed digital climate-to take shape in this particular organisation. © Proceedings of the 24th Pacific Asia Conference on information Systems: information Systems (IS) for the Future, PACIS 2020. All rights reserved.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Blind Lanes and Zebra Crossings Segmentation Network for Embedded Platforms

SSRN

引用

SSRN 2023年

作者： Gong, Tianyi Jiang, Yan He, Lingfeng Zhang, Shizhong Yan, Shicheng Li, Linfeng Liu, Jianyang School of Information Science and Technology Southwest Jiaotong University Sichuan Chengdu610031 China School of Computing and Artificial Intelligence Southwest Jiaotong University Sichuan Chengdu610031 China School of Mechanical Engineering Southwest Jiaotong University Sichuan Chengdu610031 China

In the application of robot-assisted intelligent navigation systems, real-time and accurate segmentation of blind lanes and zebra crossings are crucial for enhancing the safety of visually impaired individuals during travel. Traditional image feature-based segmentation techniques for blind lanes and zebra crossings are often susceptible to outdoor lighting interference, while deep learningbased methods face challenges such as large network size and poor real-time performance, making them difficult to deploy and apply on embedded platforms. To address these issues, this paper proposes a new lightweight segmentation network for blind lanes and zebra crossings, targeting embedded platforms. Firstly, we retrofit a motorized wheelchair with a high-definition camera and an embedded computing platform based on the Altas310 processor. We capture 207 images containing blind lanes and zebra crossings using this camera. Subsequently, we merge and label these images with those from an open-source dataset on blind lanes, resulting in a dataset comprising 407 images and 4 video segments of blind lanes and zebra crossings. Furthermore, we propose a lightweight high-precision semantic segmentation network, which utilizes an improved G-GhostNet and depth-wise separable convolution to reduce network parameters. It combines an enhanced Atrous Spatial Pyramid Pooling module and attention mechanism to enhance feature extraction at different levels, and employs a specially designed decoding module to fuse semantic information from different levels and achieve segmentation of blind lanes and zebra crossings. Deploying the model on the embedded computing platform integrated into the wheelchair, our lightweight segmentation model significantly improves the speed of segmentation, achieving 59.2 FPS. The accuracy of blind lanes and zebra crossing segmentation reaches 98.6%, demonstrating certain advantages compared to selected state-of-the-art methods. Experimental results indicate that our p

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Enhancing Ieee 1588 Ptp Security for Iiot Networks: A Lightweight Attack Detection and Mitigation Framework

SSRN

引用

SSRN 2025年

作者： Idrees, Zeba Latif, Shahid Tahir, Hira Zheng, Liron Electrical and Computer Engineering Department University of Alberta Canada School of Computing and Creative Technologies University of the West of England United Kingdom Interdisciplinary Research Centre for Smart Mobility and Logistics King Fahd University of Petroleum and Minerals Dhahran31261 Saudi Arabia School of Information Science and Technology Fudan University China

Highly precise clock synchronization is an important aspect of the Industrial Internet of Things (IIoT) network because desynchronized clocks among nodes in IIoT can degrade system performance. IEEE 1588 Precision Time Protocol (PTP) is widely used in such time-sensitive networks. Resource efficiency and security have become the most important concerns in designing PTP for IIoT applications. PTP provides unified and high-precision time, whereas it is resource inefficient and insecure in its current form, particularly for resource-constrained IoT devices, such as battery powered sensing nodes. This paper aims to advance the existing PTP to improve security for IIoT networks without involving complex and power-consuming cryptographic algorithms. We study and analyze the potential cyber-attacks that can affect the security and synchronization of the PTP network. Considering the limitations of the PTP security defined by IEEE 1588 in its Annex K, we propose a security extension to the PTP algorithm. This security model covers the full PTP attack surface and allows the detection of attacks on all the PTP nodes in a timely manner. Along with the attack detection, we establish an attack mitigation model to mitigate the attack effects on Master PTP nodes. The proposed secure PTP model was evaluated under different network conditions and with varying important parameters. It was observed that newly introduced functions do not compromise synchronization accuracy. All the experimental evaluations demonstrate that the proposed approach is more secure and robust to cyber-attacks and does not affect the operation of PTP devices in all considered network configurations. © 2025, The Authors. All rights reserved.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

Supplemental information

arXiv

引用

arXiv 2025年

作者： Sunada, Satoshi Niiyama, Tomoaki Kanno, Kazutaka Nogami, Rin Röhm, André Awano, Takato Uchida, Atsushi Faculty of Mechanical Engineering Institute of Science and Engineering Kanazawa University Kakuma-machi Kanazawa Ishikawa920-1192 Japan Department of Information and Computer Sciences Saitama University 255 Shimo-Okubo Sakura-ku Saitama338-8570 Japan Department of Information Physics and Computing Graduate School of Information Science and Technology The University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo113-8656 Japan

The rapidly increasing computational demands for artificial intelligence (AI) have spurred the exploration of computing principles beyond conventional digital computers. Physical neural networks (PNNs) offer efficient neuromorphic information processing by harnessing the innate computational power of physical processes;however, training their weight parameters is computationally expensive. We propose a training approach for substantially reducing this training cost. Our training approach merges an optimal control method for continuous-time dynamical systems with a biologically plausible training method—direct feedback alignment. In addition to the reduction of training time, this approach achieves robust processing even under measurement errors and noise without requiring detailed system information. The effectiveness was numerically and experimentally verified in an optoelectronic delay system. Our approach significantly extends the range of physical systems practically usable as PNNs. Copyright © 2025, The Authors. All rights reserved.

关键词： Continuous time systems

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：