检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Ma, Feng Wang, Xiu-Min Chen, Chen Xu, Xiao-Bin Yan, Xin-Ping Wuhan University of Technology Wuhan China Intelligent Transportation Systems Research Center Wuhan University of Technology Wuhan China School of Computer Science & Engineering Wuhan Institute of Technology Wuhan China China-Austria Belt and Road Joint Laboratory on Artificial Intelligence and Advanced Manufacturing Hangzhou Dianzi University Hangzhou China Nanjing Smart Water Transport Technology Co. Ltd. Nanjing China

Existing navigation decision support systems often perform poorly when handling non-predefined navigation scenarios. Leveraging the generalization capabilities of large language model (LLM) in handling unknown scenarios, this research proposes a dual-core framework for LLM applications to address this issue. Firstly, through ReAct-based prompt engineering, a larger LLM core decomposes intricate navigation tasks into manageable sub-tasks, which autonomously invoke corresponding external tools to gather relevant information, using this feedback to mitigate the risk of LLM hallucinations. Subsequently, a fine-tuned and compact LLM core, acting like a first-mate is designed to process such information and unstructured external data, then to generates context-aware recommendations, ultimately delivering lookout insights and navigation hints that adhere to the International Regulations for Preventing Collisions at Sea (COLREGs) and other rules. Extensive experiments demonstrate the proposed framework not only excels in traditional ship collision avoidance tasks but also adapts effectively to unstructured, non-predefined, and unpredictable scenarios. A comparative analysis with DeepSeek-R1, GPT-4o and other SOTA models highlights the efficacy and rationality of the proposed framework. This research bridges the gap between conventional navigation systems and LLMs, offering a framework to enhance safety and operational efficiency across diverse navigation applications. © 2025, CC BY-SA.

关键词： Ships

来源：评论

学校读者我要写书评

暂无评论

Effects of Air Exposure on the Interfacial Capacitance of Laser-Induced Graphene (LIG) with and without Silver (Ag) Nanoparticle Loading

Effects of Air Exposure on the Interfacial Capacitance of La...

引用

IEEE International Conference on Sensors and Nanotechnology

作者： Rou Xuan Goh Suhana Mohamed Sultan Suan Hui Pu Pei Ling Leow Hing Wah Lee Sam J. Fishlock Harold M. H. Chong Ting Yang Ling Smart Manufacturing and Systems Research Group University of Southampton Malaysia Iskandar Puteri Johor Malaysia Faculty of Electrical Engineering Universiti Teknologi Malaysia Skudai Johor Malaysia Centre for Semiconductor & Thin Film Research MIMOS Berhad Technology Park Malaysia Kuala Lumpur Malaysia The Engineering & Design Institute TEDI-London London U.K. School of Electronics and Computer Science University of Southampton Southampton U.K.

research on laser-induced graphene (LIG) and LIG with silver nanoparticles loading (Ag-LIG) are increasing by the day as more of its potential are uncovered each day. With the LIG fabrication looking to increase further in the future, it will be useful to confirm and verify any changes in the quality of LIG when it is exposed to the ambient air. In normal laboratory conditions, researchers store samples in petri dishes or any other covered storage, but inevitably the air within the storage will still be in contact with the samples. In this study, we left LIG and Ag-LIG samples in a lidded petri dish after its fabrication and measured the interfacial capacitance of each sample using a standard three-electrode electrochemical setup with a potential range from 0 V to 0.4 V on different days. Results show that the interfacial capacitance of both LIG and Ag-LIG drops as it is left in the storage container longer after its fabrication, with Ag-LIG showing a greater decrease after 10 days compared to LIG. This indicates that the silver nanoparticles played an important part during the oxidation process, making Ag-LIG more sensitive to its environment during prolonged unuse. The findings from this study shows the potential usage of LIG and Ag-LIG as a low-cost single-use indicator for air exposure.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Matching Game for Optimized Association in Quantum Communication Networks

Matching Game for Optimized Association in Quantum Communica...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Mahdi Chehimi Bernd Simon Walid Saad Anja Klein Don Towsley Mérouane Debbah Bradley Department of Electrical and Computer Engineering Wireless@VT Virginia Tech Arlington VA USA Communication Engineering Lab Technische Universität Darmstadt Darmstadt Germany Cyber Security Systems and Applied AI Research Center Lebanese American University Lebanon College of Information and Computer Sciences University of Massachusetts Amherst Amherst MA USA Electrical Engineering and Computer Science Department Khalifa University of Science and Technology Abu Dhabi UAE

Enabling quantum switches (QSs) to serve requests submitted by quantum end nodes in quantum communication networks (QCNs) is a challenging problem due to the heterogeneous fidelity requirements of the submitted requests and the limited resources of the QCN. Effectively determining which requests are served by a given QS is fundamental to foster developments in practical QCN applications, like quantum data centers. However, the state-of-the-art on QS operation has overlooked this association problem, and it mainly focused on QCNs with a single QS. In this paper, the request-QS association problem in QCNs is formulated as a matching game that captures the limited QCN resources, heterogeneous application-specific fidelity requirements, and scheduling of the different QS operations. To solve this game, a swap-stable request-QS association (RQSA) algorithm is proposed while considering partial QCN information availability. Extensive simulations are conducted to validate the effectiveness of the proposed RQSA algorithm. Simulation results show that the proposed RQSA algorithm achieves a near-optimal (within 5%) performance in terms of the percentage of served requests and overall achieved fidelity, while outperforming benchmark greedy solutions by over 13%. Moreover, the proposed RQSA algorithm is shown to be scalable and maintain its near-optimal performance even when the size of the QCN increases.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mean-field games among teams

arXiv

引用

arXiv 2023年

作者： Subramanian, Jayakumar Kumar, Akshat Mahajan, Aditya Media and Data Science Research Lab Digital Experience Cloud Adobe Inc. Uttar Pradesh Noida India School of Computing and Information Systems The Singapore Management University Singapore Department of Electrical and Computer Engineering McGill University Montreal Canada

In this paper, we present a model of a game among teams. Each team consists of a homogeneous population of agents. Agents within a team are cooperative while the teams compete with other teams. The dynamics and the costs are coupled through the empirical distribution (or the mean field) of the state of agents in each team. This mean-field is assumed to be observed by all agents. Agents have asymmetric information (also called a non-classical information structure). We propose a mean-field based refinement of the Team-Nash equilibrium of the game, which we call mean-field Markov perfect equilibrium (MF-MPE). We identify a dynamic programming decomposition to characterize MF-MPE. We then consider the case where each team has a large number of players and present a mean-field approximation which approximates the game among large-population teams as a game among infinite-population teams. We show that MF-MPE of the game among teams of infinite population is easier to compute and is an Ε-approximate MF-MPE of the game among teams of finite population. Copyright © 2023, 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

Leveraging SDN for Real World Windfarm Process Automation Architectures

Leveraging SDN for Real World Windfarm Process Automation Ar...

引用

International Conference on Software Defined systems (SDS)

作者： Soufiane Hamadache Elhadj Benkhelifa Hisham Kholidy Pradeeban Kathiravelu Brij B. Gupta General Electric. Staffordshire UK Smart Systems AI and Cybersecurity Research Centre Staffordshire University Stoke on Trent UK ACRL lab College of Engineering State University of New York Polytechnic Institue USA University of Alaska Anchorage Anchorage AK USA Asia University Taiwan Taichung Taiwan

SCADA systems are the main part of automated controls and supervision in industrial control systems (ICS). Current SCADA systems are installed in sophisticated, complex architectures involving many third-party sub systems connected to the Internet for remote access/control and system updates. The existing architectures used in windfarms are based on classical switches where everything is enabled by default with plug and play ports, and where the administrator is required to interact with the physical infrastructure to decide what to disable such as unused ports and insecure services (Telnet, http). The classical switches have limited functions and logics. Software-Defined Networking (SDN) is an eminent technology, it decouples the control and data planes, centralizes logically the network devices and abstract the network infrastructure from the application layers. Therefore, it can interact with upstream Application to allow the extending of the functionality of a programmable network by users. Protocol like IEC61850+(60870-5-104) or (60870-5-101) are commonly used in SCADA networks to operate critical infrastructures (CI), such as power plants and substations. It is the protocol used to collect data and operate the wind turbines' interfaces; however, these protocols have never been tested in SDN. Emulation environments, such as Mininet, which is the closest work reported in literature, does not reflect the real world with real devices. Therefore, the aim of this paper is to leverage SDN technology in SCADA architecture for real world Windfarm projects. Two SDN controllers were implemented and compared, namely OpenDaylight and Ryu

关键词：

来源：评论

学校读者我要写书评

暂无评论

Odour Localization in Neuromorphic systems

Odour Localization in Neuromorphic Systems

引用

IEEE International Symposium on Circuits and systems (ISCAS)

作者： Thorben Schoepe Damien Drix Franz Marcus Schüffny Rebecca Miko Samuel Sutton Elisabetta Chicca Michael Schmuker Peter Grünberg Institut 15 Forschungszentrum Jülich Aachen Germany Bio-Inspired Circuits and Systems Lab Zernike Institute for Advanced Materials University of Groningen Netherlands Groningen Cognitive Systems and Materials Center University of Groningen Netherlands Biocomputation Research Group University of Hertfordshire Hatfield United Kingdom Faculty of Electrical Engineering and Computer Science TU Dresden Dresden Germany

ISBN: (数字)9798350330991

ISBN: (纸本)9798350331004

Odour source localization is crucial in life-saving scenarios like pinpointing gas leaks, detecting explosives, searching for earthquake survivors, or locating fires at their origin. The turbulent character of natural environments makes this task very challenging. The absolute concentration of odour plumes carries little meaning and these plumes are only encountered in an intermittent, transient fashion. However, navigation algorithms that are driven by odour encounter events, can successfully find odour sources by extracting spatiotemporal information. The event driven nature of odour plumes motivates a fully event-driven sensing and processing pipeline for robot navigation. Hence, we developed a spiking neural network, implemented on neuromorphic hardware, that can successfully decode odour-puff direction from a pair of enose-systems. This is to our knowledge the first fully event driven neuromorphic system for odour localization.

关键词： Location awareness Neuromorphics Navigation Pipelines Green products Robot sensing systems Sensors

来源：评论

学校读者我要写书评

暂无评论

Adaptive Backdoor Attacks with Reasonable Constraints on Graph Neural Networks

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Dong, Xuewen Li, Jiachen Li, Shujun You, Zhichao Qu, Qiang Kholodov, Yaroslav Shen, Yulong The School of Computer Science and Technology Xidian University The Engineering Research Center of Blockchain Technology Application and Evaluation Ministry of Education China The Shaanxi Key Laboratory of Blockchain and Secure Computing Xi’an710071 China The School of Computing Kent Interdisciplinary Research Centre in Cyber Security University of Kent CanterburyCT2 7NF United Kingdom Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China The Intelligent Transportation Systems Lab Innopolis University Innopolis Russia The School of Computer Science and Technology Xidian University China

Recent studies show that graph neural networks (GNNs) are vulnerable to backdoor attacks. Existing backdoor attacks against GNNs use fixed-pattern triggers and lack reasonable trigger constraints, overlooking individual graph characteristics and rendering insufficient evasiveness. To tackle the above issues, we propose ABARC, the first Adaptive Backdoor Attack with Reasonable Constraints, applying to both graph-level and node-level tasks in GNNs. For graph-level tasks, we propose a subgraph backdoor attack independent of the graph’s topology. It dynamically selects trigger nodes for each target graph and modifies node features with constraints based on graph similarity, feature range, and feature type. For node-level tasks, our attack begins with an analysis of node features, followed by selecting and modifying trigger features, which are then constrained by node similarity, feature range, and feature type. Furthermore, an adaptive edge-pruning mechanism is designed to reduce the impact of neighbors on target nodes, ensuring a high attack success rate (ASR). Experimental results show that even with reasonable constraints for attack evasiveness, our attack achieves a high ASR while incurring a marginal clean accuracy drop (CAD). When combined with the state-of-the-art defense randomized smoothing (RS) method, our attack maintains an ASR over 94%, surpassing existing attacks by more than 7%. © 2004-2012 IEEE. 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

XAnet: Cross-Attention Between EEG of Left and Right Brain for Auditory Attention Decoding

XAnet: Cross-Attention Between EEG of Left and Right Brain f...

引用

International IEEE/EMBS Conference on Neural engineering, CNE

作者： Saurav Pahuja Siqi Cai Tanja Schultz Haizhou Li Machine Listening Lab University of Bremen Germany Department of Electrical and Computer Engineering National University of Singapore Singapore Cognitive Systems Lab University of Bremen Germany Machine Listening Lab Shenzhen Research Institute of Big Data School of Data Science The Chinese University of Hong Kong Shenzhen China Kriston AI Xiamen China

Humans have the ability to segregate different incoming sounds and perform selective attention to a particular sound using the brain's auditory attention mechanism. Various psycho-acoustic studies suggest that acoustic features extracted by both brain hemispheres are integrated through interhemispheric integration. During selective auditory attention, different responses are then present in the left and right sides of the brain. Recently, studies also suggest that attention induces a spatially specific alpha power increase and decrease in respective hemispheres ipsilateral and contralateral to the locus of attended speech. This finding motivates us to design XAnet, a neuroscience-inspired network for Auditory Attention Decoding (AAD). In XAnet, neural activity measured by ElectroEncephalography (EEG) is first divided into left and right-side EEG channel groups and then integrated as cross-attention to capture differentiated left and right-side responses of the brain. To the best of our knowledge, this is the first approach in AAD that divides and integrates the left and right channel groups to capture brain responses. Our experimental results indicate that XAnet outper-forms other competitive models. It achieves (90.6%, 82.1%, and 75.1%) accuracy for 1-second decision window and outperformed the benchmark CNN (84.1%, 75.1%, and 62.0%) on the KUL dataset for 64, 10, and 4 channels, respectively.

关键词： Neural activity Neural engineering Benchmark testing Feature extraction Brain modeling Electroencephalography Acoustics

来源：评论

学校读者我要写书评

暂无评论

Searching for Network Width with Bilaterally Coupled Network

arXiv

引用

arXiv 2022年

作者： Su, Xiu You, Shan Xie, Jiyang Wang, Fei Qian, Chen Zhang, Changshui Xu, Chang School of Computer Science Faculty of Engineering The University of Sydney Australia Pattern Recognition and Intelligent Systems Lab. Beijing University of Posts and Telecommunications China SenseTime Research Centre China

Searching for a more compact network width recently serves as an effective way of channel pruning for the deployment of convolutional neural networks (CNNs) under hardware constraints. To fulfill the searching, a one-shot supernet is usually leveraged to efficiently evaluate the performance w.r.t. different network widths. However, current methods mainly follow a unilaterally augmented (UA) principle for the evaluation of each width, which induces the training unfairness of channels in supernet. In this paper, we introduce a new supernet called Bilaterally Coupled Network (BCNet) to address this issue. In BCNet, each channel is fairly trained and responsible for the same amount of network widths, thus each network width can be evaluated more accurately. Besides, we propose to reduce the redundant search space and present the BCNetV2 as the enhanced supernet to ensure rigorous training fairness over channels. Furthermore, we leverage a stochastic complementary strategy for training the BCNet, and propose a prior initial population sampling method to boost the performance of the evolutionary search. We also propose the first open-source width benchmark on macro structures named Channel-Bench-Macro for the better comparison of width search algorithms. Extensive experiments on benchmark CIFAR-10 and ImageNet datasets indicate that our method can achieve state-of-the-art or competing performance over other baseline methods. Moreover, our method turns out to further boost the performance of NAS models by refining their network widths. For example, with the same FLOPs budget, our obtained EfficientNet-B0 achieves 77.53% Top-1 accuracy on ImageNet dataset, surpassing the performance of original setting by 0.65%. Copyright © 2022, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

smartEdge: smart Healthcare End-to-End Integrated Edge and Cloud Computing System for Diabetes Prediction Enabled by Ensemble Machine Learning

arXiv

引用

arXiv 2025年

作者： Hennebelle, Alain Dieng, Qifan Ismail, Leila Buyya, Rajkumar Abu Dhabi United Arab Emirates Lab School of Computing and Information Systems University of Melbourne Melbourne Australia Lab Department of Computer Science and Software Engineering College of Information Technology United Arab Emirates Emirates Center for Mobility Research United Arab Emirates National Water and Energy Center UAE University Abu Dhabi Al-Ain United Arab Emirates

The Internet of Things (IoT) revolutionizes smart city domains such as healthcare, transportation, industry, and education. The Internet of Medical Things (IoMT) is gaining prominence, particularly in smart hospitals and Remote Patient Monitoring (RPM). The vast volume of data generated by IoMT devices should be analyzed in real-time for health surveillance, prognosis, and prediction of diseases. Current approaches relying on Cloud computing to provide the necessary computing and storage capabilities do not scale for these latency-sensitive applications. Edge computing emerges as a solution by bringing cloud services closer to IoMT devices. This paper introduces smartEdge, an AI-powered smart healthcare end-to-end integrated edge and cloud computing system for diabetes prediction. This work addresses latency concerns and demonstrates the efficacy of edge resources in healthcare applications within an end-to-end system. The system leverages various risk factors for diabetes prediction. We propose an Edge and Cloud-enabled framework to deploy the proposed diabetes prediction models on various configurations using edge nodes and main cloud servers. Performance metrics are evaluated using, latency, accuracy, and response time. By using ensemble machine learning voting algorithms we can improve the prediction accuracy by 5% versus a single model *** Codes 68T01, 68T09, 68M14, 68W10, 68W15 © 2025, CC BY.

关键词： Electronic health record

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：