检索结果-内蒙古大学图书馆

Tensor Levenberg-Marquardt Algorithm for Multi-Relational Traffic Prediction

IEEE Transactions on Vehicular Technology 2023年第9期72卷 11275-11290页

作者： Chang, Shih Yu Wu, Hsiao-Chun Kuan, Yen-Cheng Wu, Yiyan San Jose State University Department of Applied Data Science San JoseCA95192 United States Louisiana State University School of Electrical Engineering and Computer Science Baton RougeLA70803 United States Yuan Ze University Innovation Center for AI Applications Chungli32003 Taiwan International College of Semiconductor Technology Hsinchu30010 Taiwan Communications Research Centre Canada OttawaONK2H 8S2 Canada

Traffic prediction serves as a crucial role in intelligent transportation systems for applications including routing planning and traffic control. With the advances of sensor technologies and Internet of Things (IoTs), the exploitation of traffic data motivates us to reformulate the traffic-state prediction problem based on the big-data characteristics. In order to infer multifarious traffic information, the standard one/two-relational traffic data in the vector/matrix form needs to be expanded to multi-relational traffic data in an arbitrary tensor form. However, none of the existing approaches is capable of performing traffic prediction by characterizing and tracking the inherent multi-relational nonlinear dynamics which are often encountered in the realistic time-series analysis. Furthermore, there hardly exist convergence studies on convergence condition and convergence speed pertinent to the existing traffic-prediction methods. In order to combat the aforementioned challenges, we propose a new tensor Levenberg-Marquadt algorithm (TLMA) to conduct the multi-relational nonlinear regression for traffic prediction involving tensorized data. Meanwhile, we also derive the lower-bound of the number of iterations required to make the error-norm below a specified tolerance for global convergence. Besides, we derive the conditions for our proposed TLMA to converge to the optimal solution. The convergence rates with respect to different nonlinear cost functions have been derived theoretically. The computational- and memory-complexities of our proposed new TLMA are studied as well. Finally, numerical experiments are conducted to evaluate the traffic prediction performance of our proposed new TLMA over real traffic data in comparison with other existing multi-relational prediction methods. © 1967-2012 IEEE.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Energy-Aware Task Allocation for Teams of Multi-mode Robots

arXiv

引用

arXiv 2025年

作者： Ito, Takumi Funada, Riku Sampei, Mitsuji Notomista, Gennaro Department of Systems and Control Engineering Institute of Science Tokyo Tokyo152-8550 Japan Department of Electrical and Computer Engineering University of Waterloo WaterlooONN2L 3G1 Canada

This work proposes a novel multi-robot task allocation framework for robots that can switch between multiple modes, e.g., flying, driving, or walking. We first provide a method to encode the multi-mode property of robots as a graph, where the mode of each robot is represented by a node. Next, we formulate a constrained optimization problem to decide both the task to be allocated to each robot as well as the mode in which the latter should execute the task. The robot modes are optimized based on the state of the robot and the environment, as well as the energy required to execute the allocated task. Moreover, the proposed framework is able to encompass kinematic and dynamic models of robots alike. Furthermore, we provide sufficient conditions for the convergence of task execution and allocation for both robot models. Copyright © 2025, The Authors. All rights reserved.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Outage, Capacity, and Error Performance of Downlink RSMA-based Systems: Analysis and Resource Optimization

引用

IEEE Transactions on Communications 2025年

作者： Vu, Thai-Hoc Da Costa, Daniel Benevides Kim, Sunghwan Pham, Quoc-Viet University of Ulsan Department of Electrical Electronic and Computer Engineering Ulsan Korea Republic of Interdisciplinary Research Center for Communication Systems and Sensing Department of Electrical Engineering Dhahran31261 Saudi Arabia Kyonggi University School of Electronic Engineering Kyonggi Korea Republic of University of Dublin School of Computer Science and Statistics Trinity College Dublin Dublin 2 D02PN40 Ireland

This paper comprehensively investigates the performance of downlink multi-user rate-splitting multiple access (RSMA) networks under Nakagami-m fading channels. We first develop the mathematical outage probability (OP) and ergodic capacity (EC) frameworks, deriving exact expressions for both, along with asymptotic analysis in high signal-to-noise ratio (SNR) and low-rate regions, which serve as the foundation for deducing the approximate and maximal energy-reliability and energy-spectral formulas. To enhance system performance, we tackle the non-convex problems of jointly optimizing common and private power allocation (PA) coefficients to minimize the maximal OP performance. Moreover, we also delve into optimizing PA coefficients and rate-splitting factors concurrently to maximize the ergodic sum capacity (ESC). Addressing the influence of modulation schemes on user error rates, we introduce mathematical frameworks for symbol error rate (SER) considering four combined modulation schemes based on binary phase-shift keying (BPSK) and quadrature-phase shift keying (QPSK), the whole cases are quantified in terms of exact and asymptotic manners. Furthermore, we present a straightforward approach to optimize the PA coefficients to minimize the maximal SER performance. Finally, Monte-Carlo simulations are presented to validate our developed frameworks and optimization solutions. © 1972-2012 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Explainable Label-flipping Attacks on Human Emotion Assessment System

arXiv

引用

arXiv 2023年

作者： Zhang, Zhibo Hammadi, Ahmed Al Y. Damiani, Ernesto Yeun, Chan Yeob Yoon, Sangyoung C2PS Department of Electrical Engineering and Computer Science Khalifa University Abu Dhabi United Arab Emirates

This paper's main goal is to provide an attacker's point of view on data poisoning assaults that use label-flipping during the training phase of systems that use electroencephalogram (EEG) signals to evaluate human emotion. To attack different machine learning classifiers such as Adaptive Boosting (AdaBoost) and Random Forest dedicated to the classification of 4 different human emotions using EEG signals, this paper proposes two scenarios of label-flipping methods. The results of the studies show that the proposed data poison attacks based on label-flipping are successful regardless of the model, but different models show different degrees of resistance to the assaults. In addition, numerous Explainable Artificial Intelligence (XAI) techniques are used to explain the data poison attacks on EEG signal-based human emotion evaluation systems. © 2023, CC BY.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Timed I/O Automata: It Is Never Too Late to Complete Your Timed Specification Theory

arXiv

引用

arXiv 2023年

作者： Goorden, Martijn A. Larsen, Kim G. Legay, Axel Lorber, Florian Nyman, Ulrik Wąsowski, Andrzej Department of Computer Science Aalborg University Selma Lagerlöfs Vej 300 Öst Aalborg9220 Denmark Department of Computer Science Engineering UC Louvain Place Sainte Barbe 2 Louvain-la-Neuve1348 Belgium Department of Computer Science IT University Rued Langgaards Vej 7 Copenhagen S2300 Denmark

A specification theory combines notions of specifications and implementations with a satisfaction relation, a refinement relation and a set of operators supporting stepwise design. We develop a complete specification framework for real-time systems using Timed I/O Automata as the specification formalism, with the semantics expressed in terms of Timed I/O Transition Systems. We provide constructs for refinement, consistency checking, logical and structural composition, and quotient of specifications—all indispensable ingredients of a compositional design methodology. The theory is backed by rigorous proofs and is being implemented in the open-source tool ECDAR. © 2023, CC BY.

关键词： Specifications

来源：评论

学校读者我要写书评

暂无评论

Analyzing and Comparing Results for Multiple Machine Learning Models 4

Analyzing and Comparing Results for Multiple Machine Learnin...

引用

2023 4th International Conference on computer engineering and Intelligent Control, ICCEIC 2023

作者： Ng, Jefferson Li, Yijian Ji, Xuxiao Yang, Can Wan, Jiaqiao Lu, Xingyu University of California Department of Mechanical Engineering Irvine92697 United States Courant Institute of Mathematical Sciences New York University NY10012 United States University of Waterloo School of Computer Science WaterlooCAN2L 3G1 United States Oregon State University Department of Electrical Engineering and Computer Science Corvallis97331 United States Zhengzhou University Software College Zhengzhou450000 China United World College Changshu China Jiangsu215500 China

ISBN: (纸本)9798350308877

In the contemporary landscape of atmospheric sciences, the ability to recognize specific weather phenomena accurately and efficiently has become paramount, largely due to the exponential growth in sensor-generated data. This paper undertakes an in-depth analysis of advanced algorithms, namely YOLOV8, ResNet50, and Convolutional Neural Networks (CNN), for the purpose of weather pattern identification and classification. Building on the foundation of deep learning, the YOLOV8's real-time object detection capabilities are leveraged to discern intricate weather patterns in diverse datasets, from meteorological stations to satellite imagery. On the other hand, Resnet 50 which is another CNN based model was also tested along with CNN for comparison. An exhaustive evaluation of these algorithms covers various metrics, including accuracy, precision, computational efficiency, and real-world applicability. Special emphasis is placed on the crucial stages of data preprocessing, feature extraction, and model tuning, highlighting their impact on the algorithms' overall performance. Our findings suggest that, when appropriately optimized, both YOLOV8 and CNN exhibit exceptional capabilities in discerning and classifying intricate weather patterns, whereas ResNet50 exhibits comparatively less performance. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Load analysis on the drone protection cage increasing collision resistance

引用

Procedia Structural Integrity 2024年 54卷 506-513页

作者： Wojciech Skarka Magdalena Szczepanek Maciej Pośpiech Aleksander Jassak Jakub Żymełka Michał Pokrzywa Michał Górka Roman Niestrój Department of Fundamentals of Machinery Design Silesian University of Technology Konarskiego 18A 44-100 Gliwice Poland Department of Electrical Engineering and Computer Science Silesian University of Technology Akademicka 2A 44-100 Gliwice Poland

Drones are currently being used in more and more new fields, including those where there are many obstacles that a drone can hit while performing its task. Therefore, the need to prevent a drone from damage when it crashes into an obstacle has arisen and such protection will also enable the use of drones in new applications involving operating in a very small space with a high probability of collision. One of the solutions to this problem is to put the drone into a protective cage. This paper presents the results of load analysis on drone protective cage and the proposed one has a specific structure and consists of an external mesh with specially shaped nodes and topology of mesh elements. The whole cage is suspended on a gimbal type connection that increases the collision capabilities and minimizes the impact of collisions on the drone's behaviour and cage load. The unusual construction of the drone itself, and in particular the arrangement of the drone's propellers, allows us to maximize the operational possibilities. he results show how the loads are distributed throughout the cage and gimbal element and how they affect the drone inside. The tests were carried out both in a simulation environment and on built demonstrators in laboratory conditions and simulated operational conditions.

关键词： Load analysis Protection cage Collision resistance

来源：评论

学校读者我要写书评

暂无评论

AI-aided power electronic converters automatic online real-time efficiency optimization method

引用

Fundamental Research 2025年第3期5卷 1111-1116页

作者： Tang, Yuanhong Cao, Di Xiao, Jian Jiang, Chenying Huang, Qi Li, Yunwei Chen, Zhe Blaabjerg, Frede Hu, Weihao Power System Wide-area Measurement and Control Sichuan Provincial Key Laboratory School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu 611731 China MORNSUN Guangzhou Science & Technology Co. Ltd. Guangzhou 510663 China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu 611731 China Yangtze Delta Region Institute (Quzhou) University of Electronic Science and Technology of China Quzhou 324000 China Department of Chengdu State Grid Corporation of China Chengdu 610599 China Southwest University of Science and Technology Mianyang 621010 China Department of Electrical and Computer Engineering University of Alberta Edmonton AB T6G 2V4 Canada Department of Energy Technology Aalborg University Aalborg 9220 Denmark

Energy losses during the conversion and supply of electric power are considered a significant issue and cannot be estimated. Improvement in the efficiency of energy conversion systems is highly restricted because of their internal nonlinearity and complexity. Thus, inspired by the successful utilization of robotic chemists, we demonstrate a pioneering concept of artificial intelligence (AI)-aided automatic online real-time optimization of a power electronics converter using a dual active bridge (DAB) converter as an example. An optimal modulation strategy was obtained through repeated automatic exploration experiments on a practical DAB converter platform. Specifically, the DAB experimental platform operated autonomously around the clock for approximately 71 h. It performed 120,000 consecutive experiments (12,000 episodes) within a six-variable experimental space driven by a deep deterministic policy gradient (DDPG) algorithm. The proposed AI-aided automatic online real-time optimization method achieved significantly improved efficiency of power conversion and supply. Consequently, zero carbon emissions may be obtained in the future. © 2023

关键词： Artificial intelligence Automatic online real-time exploration DC-DC converter Dual active bridge converter Energy conversion system Power electronics

来源：评论

学校读者我要写书评

暂无评论

Leveraging Deep Reinforcement Learning for Geolocation-based MIMO Transmission in FD-RAN

Leveraging Deep Reinforcement Learning for Geolocation-based...

引用

2023 IEEE/CIC International Conference on Communications in China, ICCC 2023

作者： Liu, Zongxi Yu, Kai Zhang, Tianqi Liu, Jingbo Chen, Jiacheng Zhou, Haibo Shen, Xuemin Sherman Nanjing University School of Electronic Science and Engineering Nanjing210023 China Peng Cheng Laboratory Department of Mathematics and Theories Shenzhen518000 China Shanghai Jiao Tong University School of Electronic Information and Electrical Engineering Shanghai200240 China University of Waterloo Department of Electrical and Computer Engineering WaterlooN2L 3G1 Canada

ISBN: (纸本)9798350345384

In the fully-decoupled radio access network (FD-RAN), the base stations (BSs) are classified as control BSs, uplink BSs, and downlink BSs for better flexibility of the BS-user equipment (UE) association and elasticity of resource scheduling. However, the high feedback delay brought by the physically decoupling architecture renders conventional feedback-based MIMO transmission ineffective. Considering the strong correlation between user geolocation and the channel propagation environment, especially when a line-of-sight (LOS) path exists, we propose a geolocation-based MIMO transmission scheme that determines the user's precoder and modulation and coding scheme (MCS) without requiring the conventional feedback procedures. To capture the correlation, we utilize deep reinforcement learning (DRL) to perform the task of mapping from geolocation to transmission parameters. Based on the principles of Type I codebook, we further decompose the task into subtasks to reduce the action space. The extensive simulations on realistic ray-tracing channels demonstrate comparable performance to CSI feedback-based methods in 5G under ideal (zero feedback delay) conditions and better performance against 5G (with feedback delay) and DRL without task decomposition. © 2023 IEEE.

关键词： Radio access networks

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Collaboration Mechanisms: A Survey of LLMs

arXiv

引用

arXiv 2025年

作者： Tran, Khanh-Tung Dao, Dung Nguyen, Minh-Duong Pham, Quoc-Viet O’Sullivan, Barry Nguyen, Hoang D. School of Computer Science and Information Technology University College Cork Cork Ireland Department of Information Convergence Engineering Pusan National University Busan Korea Republic of School of Computer Science and Statistics Trinity College Dublin Dublin 2 D02PN40 Ireland

With recent advances in Large Language Models (LLMs), Agentic AI has become phenomenal in real-world applications, moving toward multiple LLM-based agents to perceive, learn, reason, and act collaboratively. These LLM-based Multi-Agent Systems (MASs) enable groups of intelligent agents to coordinate and solve complex tasks collectively at scale, transitioning from isolated models to collaboration-centric approaches. This work provides an extensive survey of the collaborative aspect of MASs and introduces an extensible framework to guide future research. Our framework characterizes collaboration mechanisms based on key dimensions: actors (agents involved), types (e.g., cooperation, competition, or coopetition), structures (e.g., peer-to-peer, centralized, or distributed), strategies (e.g., role-based or model-based), and coordination protocols. Through a review of existing methodologies, our findings serve as a foundation for demystifying and advancing LLM-based MASs toward more intelligent and collaborative solutions for complex, real-world use cases. In addition, various applications of MASs across diverse domains, including 5G/6G networks, Industry 5.0, question answering, and social and cultural settings, are also investigated, demonstrating their wider adoption and broader impacts. Finally, we identify key lessons learned, open challenges, and potential research directions of MASs towards artificial collective intelligence. Copyright © 2025, The Authors. All rights reserved.

关键词：

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：