检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wu, Yinyu Zhang, Xuhui Ren, Jinke Shen, Yanyan Yang, Bo Wang, Shuqiang Guan, Xinping Niyato, Dusit The Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong518055 China The University of Chinese Academy of Sciences Beijing100049 China The Shenzhen Future Network of Intelligence Institute The School of Science and Engineering The Guangdong Provincial Key Laboratory of Future Networks of Intelligence The Chinese University of Hong Kong Guangdong Shenzhen518172 China Shenzhen University of Advanced Technology Guangdong518055 China The Department of Automation The Key Laboratory of System Control and Information Processing Ministry of Education Shanghai Jiao Tong University Shanghai200240 China The College of Computing and Data Science Nanyang Technological University Singapore639798 Singapore

Intelligent reflecting surface (IRS)-assisted mobile edge computing (MEC) systems have shown notable improvements in efficiency, such as reduced latency, higher data rates, and better energy efficiency. However, the resource competition among users will lead to uneven allocation, increased latency, and lower throughput. Fortunately, the rate-splitting multiple access (RSMA) technique has emerged as a promising solution for managing interference and optimizing resource allocation in MEC systems. This paper studies an IRS-assisted MEC system with RSMA, aiming to jointly optimize the passive beamforming of the IRS, the active beamforming of the base station, the task offloading allocation, the transmit power of users, the ratios of public and private information allocation, and the decoding order of the RSMA to minimize the average delay from a novel uplink transmission perspective. Since the formulated problem is non-convex and the optimization variables are highly coupled, we propose a hierarchical deep reinforcement learning-based algorithm to optimize both continuous and discrete variables of the problem. Additionally, to better extract channel features, we design a novel network architecture within the policy and evaluation networks of the proposed algorithm, combining convolutional neural networks and densely connected convolutional network for feature extraction. Simulation results indicate that the proposed algorithm not only exhibits excellent convergence performance but also outperforms various benchmarks. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Strategy evolution on dynamic networks

arXiv

引用

arXiv 2023年

作者： Su, Qi McAvoy, Alex Plotkin, Joshua B. Department of Automation Shanghai Jiao Tong University Shanghai200240 China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China Shanghai Engineering Research Center of Intelligent Control and Management Shanghai200240 China School of Data Science and Society University of North Carolina at Chapel Hill Chapel HillNC27599 United States Department of Mathematics University of North Carolina at Chapel Hill Chapel HillNC27599 United States Department of Biology University of Pennsylvania PhiladelphiaPA19104 United States Center for Mathematical Biology University of Pennsylvania PhiladelphiaPA19104 United States

Models of strategy evolution on static networks help us understand how population structure can promote the spread of traits like cooperation. One key mechanism is the formation of altruistic spatial clusters, where neighbors of a cooperative individual are likely to reciprocate, which protects prosocial traits from exploitation. But most real-world interactions are ephemeral and subject to exogenous restructuring, so that social networks change over time. Strategic behavior on dynamic networks is difficult to study, and much less is known about the resulting evolutionary dynamics. Here, we provide an analytical treatment of cooperation on dynamic networks, allowing for arbitrary spatial and temporal heterogeneity. We show that transitions among a large class of network structures can favor the spread of cooperation, even if each individual social network would inhibit cooperation when static. Furthermore, we show that spatial heterogeneity tends to inhibit cooperation, whereas temporal heterogeneity tends to promote it. Dynamic networks can have profound effects on the evolution of prosocial traits, even when individuals have no agency over network structures. © 2023, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Depth-Aware Multi-Modal Fusion for Generalized Zero-Shot Learning

Depth-Aware Multi-Modal Fusion for Generalized Zero-Shot Lea...

引用

IEEE International Conference on Industrial Informatics (INDIN)

作者： Weipeng Cao Xuyang Yao Zhiwu Xu Yinghui Pan Yixuan Sun Dachuan Li Bohua Qiu Muheng Wei Guangdong Laboratory of Artificial Intelligence and Digital Economy (Shenzhen) Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Stony Brook University New York United States Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen China Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China ZhenDui Industry Artificial Intelligence Co. Ltd Shenzhen China Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai China

ISBN: (数字)9798331527471

ISBN: (纸本)9798331527488

Realizing Generalized Zero-Shot Learning (GZSL) based on large models is emerging as a prevailing trend. However, most existing methods merely regard large models as black boxes, solely leveraging the features output by the final layer while disregarding potential performance enhancements from other layers. Indeed, numerous researchers have visually depicted variations in the features learned across different layers of neural networks. Motivated by this observation, we propose a Vision Transformer (ViT)-based GZSL method named Depth-Aware Multi-Modal ViT (DAM2ViT), which exploits multi-level features of ViT. DAM2ViT incorporates a multi-modal interaction block to align semantic information of categories across multiple layers, thereby augmenting the model's capacity to learn associations between visual and semantic spaces. Extensive experiments conducted on three benchmark datasets (i.e., CUB, SUN, AWA2) have showcased that DAM2ViT achieves competitive results compared to state-of-the-art methods.

关键词： Visualization Adaptation models Semantics Zero shot learning Neural networks Termination of employment Transformers Market research Sun Optimization

来源：评论

学校读者我要写书评

暂无评论

Designing multiple pid controllers based on an fpga for controlling the temperature of tem-cell surfaces

Designing multiple pid controllers based on an fpga for cont...

引用

2019 International Multi-Conference on Engineering, Computer and Information Sciences, SIBIRCON 2019

作者： Dubreuil, Vincent Osintsev, Artem V. School of Engineering ENSEIRB-MATMECA National Polytechnic Institute INP Bordeaux France Tomsk State University of Control Systems and Radioelectronics TUSUR Department of Data Processing Automation Tomsk Russia

ISBN: (纸本)9781728144016

A field programmable gate array (FPGA) has been configured to control the temperature average value and spatial uniformity of a TEM-cell. The system is composed of 24 Peltier elements (PE) and 52 temperature sensors (TS) distributed all around the TEM-cell. The FPGA controls each PE independently computing a proportional-integral-derivative (PID) controller with 24 different sets of parameters. The 52 TS are read using 18 Inter-Integrated-Circuit buses wired to a single master controller. This allows running the PID controller for the 24 sub-systems at a maximum frequency of 2 kHz. To compute this, a custom programmable architecture has been designed to carry out calculations with 36-bit signed values with a fixed point. The final design uses around 1600 logic elements (one logic element = one register and one 4-input lookup table), some multipliers, and some memory bits. The temperature precision obtained is around 0.1°C. The component used is an EP2C5T144C8 configured in a hardware description language Verilog with the Quartus software. © 2019 IEEE.

关键词： Pulse width modulation

来源：评论

学校读者我要写书评

暂无评论

Hexagonally ordered microbowl arrays decorated with ultrathin CuInS_(2) nanosheets for enhanced photoelectrochemical performance

引用

Journal of Energy Chemistry 2020年第12期29卷 134-142页

作者： Ming Li Le Chen Yanjie Su Huan Yin Kexiang Hu State Key Laboratory of Transducer Technology Shanghai Institute of Microsystem and Information TechnologyChinese Academy of SciencesShanghai 200050China School of Electronic Engineering and Automation Guilin University of Electronic TechnologyKey Laboratory of Optoelectronic Information Processing of Guangxi Colleges and Uniy/ersitiesGuilin 541004GuangxiChina Key Laboratory for Thin Film and Microfabrication of the Ministry of Education Department of Micro/Nano ElectronicsSchool of ElectronicsInformation and Electrical EngineeringShanghai Jiao Tong UniversityShanghai 200240China National and Local Joint Engineering Laboratory of Internet Technology on Mine School of Information and Control EngineeringChina University of Mining and TechnologyXuzhou 221008JiangsuChina School of Physics and Telecommunication Engineering Guangxi Colleges and Universities Key Lab of Complex System Optimization and Big Data ProcessingYulin Normal UniversityYulin 537400GuangxiChina

This paper demonstrates the design and fabrication of three-dimensional(3 D) hexagonally ordered microbowl arrays(MBAs) decorated with Cu In S2 nanosheets for enhanced photoelectrochemical(PEC) performance. The 3 D MBAs are fabricated by a micro-fabrication technique. The ultrathin CuInS2 nanosheets are grown on the 3 D electrodes by solvothermal transformation of Cu film. The photocurrent density of 3 D photocathode(CuInS2@MBAs) is about two times higher than that of the planar counterpart(CuInS2@Planar). The improved PEC performance can be ascribed to the elevated light trapping ability and the increased surface area for loading photocatalysts. In addition, CdS quantum dots as cocatalysts are modified onto the Cu In S2 nanosheets to further enhance the PEC activity because the formed p-n heterojunction can accelerate the separation of photogenerated carriers. As a result, the 3 D photocathode of CuInS2/CdS@MBAs shows an optimal incident photon to current efficiency of 10% at the wavelength of400 nm. It is believed that this work can be generalized to design other hierarchical 3 D photoelectrodes for improved solar water splitting.

关键词： Microbowl array CuInS_(2) Nanosheet Photocathode Photoelectrochemical performance

来源：评论

学校读者我要写书评

暂无评论

Conspiracy vs science: A large-scale analysis of online discussion cascades

arXiv

引用

arXiv 2020年

作者： Zhang, Yafei Wang, Lin Zhu, Jonathan J.H. Wang, Xiaofan Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China Department of Media and Communication School of Data Science City University of Hong Kong Hong Kong Department of Automation Shanghai University Shanghai200444 China

With the emergence and rapid proliferation of social media platforms and social networking sites, recent years have witnessed a surge of misinformation spreading in our daily life. Drawing on a large-scale dataset which covers more than 1.4M posts and 18M comments from an online social media platform, we investigate the propagation of two distinct narratives-(i) conspiracy information, whose claims are generally unsubstantiated and thus referred as misinformation to some extent, and (ii) scientific information, whose origins are generally readily identifiable and verifiable. We find that conspiracy cascades tend to propagate in a multigenerational branching process whereas science cascades are more likely to grow in a breadth-first manner. Specifically, conspiracy information triggers larger cascades, involves more users and generations, persists longer, and is more viral and bursty than science information. Content analysis reveals that conspiracy cascades contain more negative words and emotional words which convey anger, fear, disgust, surprise and trust. We also find that conspiracy cascades are much more concerned with political and controversial topics. After applying machine learning models, we achieve an AUC score of nearly 90% in discriminating conspiracy from science narratives using the constructed features. Copyright © 2020, The Authors. All rights reserved.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Image reconstruction from photoacoustic projections

引用

Photonics Insights 2024年第3期3卷 145-209页

作者： Chao Tian Kang Shen Wende Dong Fei Gao Kun Wang Jiao Li Songde Liu Ting Feng Chengbo Liu Changhui Li Meng Yang Sheng Wang Jie Tian Institute of Artificial Intelligence Hefei Comprehensive National Science CenterHefeiChina School of Engineering Science University of Science and Technology of ChinaHefeiChina Department of Anesthesiology the First Affiliated Hospital of USTCDivision of Life Sciences and MedicineUniversity of Science and Technology of ChinaHefeiChina Anhui Province Key Laboratory of Biomedical Imaging and Intelligent Processing HefeiChina College of Automation Engineering Nanjing University of Aeronautics and AstronauticsNanjingChina School of Information Science and Technology ShanghaiTech UniversityShanghaiChina Shanghai Clinical Research and Trial Center ShanghaiChina CAS Key Laboratory of Molecular Imaging Institute of AutomationChinese Academy of SciencesBeijingChina School of Precision Instruments and Optoelectronics Engineering Tianjin UniversityTianjinChina Academy for Engineering and Technology Fudan UniversityShanghaiChina Shenzhen Institute of Advanced Technology Chinese Academy of SciencesShenzhenChina Department of Biomedical Engineering College of Future TechnologyPeking UniversityBeijingChina National Biomedical Imaging Center Peking UniversityBeijingChina Departments of Ultrasound State Key Laboratory of Complex Severe and Rare DiseasesPeking Union Medical College HospitalChinese Academy of Medical Sciences and Peking Union Medical CollegeBeijingChina School of Engineering Medicine Beihang UniversityBeijingChina Key Laboratory of Big Data-Based Precision Medicine Beihang UniversityMinistry of Industry and Information TechnologyBeijingChina

Photoacoustic computed tomography(PACT)is a rapidly developing biomedical imaging modality and has attracted substantial attention in recent *** reconstruction from photoacoustic projections plays a critical role in image formation in *** we review six major classes of image reconstruction approaches developed in the past three decades,including delay and sum,filtered back projection,series expansion,time reversal,iterative reconstruction,and deep-learning-based *** principal ideas and implementations of the algorithms are summarized,and their reconstruction performances under different imaging scenarios are *** challenges,future directions,and perspectives for the development of image reconstruction algorithms in PACT are also *** review provides a self-contained reference guide for beginners and specialists in the photoacoustic community,to facilitate the development and application of novel photoacoustic image reconstruction algorithms.

关键词： optical imaging photoacoustic tomography image reconstruction deep learning

来源：评论

学校读者我要写书评

暂无评论

Confidence-Regulated Generative Diffusion Models for Reliable AI Agent Migration in Vehicular Metaverses

arXiv

引用

arXiv 2025年

作者： Kang, Yingkai Kang, Jiawen Wen, Jinbo Zhang, Tao Yang, Zhaohui Niyato, Dusit Zhang, Yan School of Automation Guangdong University of Technology Guangzhou510006 China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing210016 China School of Cyberspace Science and Technology Beijing Jiaotong University Beijing100044 China College of Information Science and Electronic Engineering Zhejiang Provincial Key Lab of information processing communication and networking Zhejiang University Hangzhou310007 China College of Computing and Data Science Nanyang Technological University Singapore Department of Informatics University of Oslo the Simula Research Laboratory Norway

Vehicular metaverses are an emerging paradigm that merges intelligent transportation systems with virtual spaces, leveraging advanced digital twin and Artificial Intelligence (AI) technologies to seamlessly integrate vehicles, users, and digital environments. In this paradigm, vehicular AI agents are endowed with environment perception, decision-making, and action execution capabilities, enabling real-time processing and analysis of multi-modal data to provide users with customized interactive services. Since vehicular AI agents require substantial resources for real-time decision-making, given vehicle mobility and network dynamics conditions, the AI agents are deployed in RoadSide Units (RSUs) with sufficient resources and dynamically migrated among them. However, AI agent migration requires frequent data exchanges, which may expose vehicular metaverses to potential cyber attacks. To this end, we propose a reliable vehicular AI agent migration framework, achieving reliable dynamic migration and efficient resource scheduling through cooperation between vehicles and RSUs. Additionally, we design a trust evaluation model based on the theory of planned behavior to dynamically quantify the reputation of RSUs, thereby better accommodating the personalized trust preferences of users. We then model the vehicular AI agent migration process as a partially observable markov decision process and develop a Confidence-regulated Generative Diffusion Model (CGDM) to efficiently generate AI agent migration decisions. Numerical results demonstrate that the CGDM algorithm significantly outperforms baseline methods in reducing system latency and enhancing robustness against cyber attacks. Copyright © 2025, The Authors. All rights reserved.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

Low-latency federated learning over wireless channels with differential privacy

arXiv

引用

arXiv 2021年

作者： Wei, Kang Li, Jun Ma, Chuan Ding, Ming Chen, Cailian Jin, Shi Han, Zhu Poor, H. Vincent School of Electrical and Optical Engineering Nanjing University of Science and Technology Nanjing210094 China Data61 CSIRO SydneyNSW2015 Australia The Department of Automation Shanghai Jiao Tong University Shanghai200240 China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China The National Mobile Communications Research Laboratory Southeast University Nanjing210096 China The Department of Electrical and Computer Engineering The University of Houston HoustonTX77004 United States The Department of Computer Science and Engineering Kyung Hee University Seoul Korea Republic of The Department of Electrical and Computer Engineering Princeton University PrincetonNJ08544 United States

In federated learning (FL), model training is distributed over clients and local models are aggregated by a central server. The performance of uploaded models in such situations can vary widely due to imbalanced data distributions, potential demands on privacy protections, and quality of transmissions. In this paper, we aim to minimize FL training delay over wireless channels, constrained by overall training performance as well as each client's differential privacy (DP) requirement. We solve this problem in the framework of multi-agent multiarmed bandit (MAMAB) to deal with the situation where there are multiple clients confornting different unknown transmission environments, e.g., channel fading and interferences. Specifically, we first transform the long-term constraints on both training performance and each client's DP into a virtual queue based on the Lyapunov drift technique. Then, we convert the MAMAB to a max-min bipartite matching problem at each communication round, by estimating rewards with the upper confidence bound (UCB) approach. More importantly, we propose two efficient solutions to this matching problem, i.e., modified Hungarian algorithm and greedy matching with a better alternative (GMBA), in which the first one can achieve the optimal solution with a high complexity while the second one approaches a better tradeoff by enabling a verified low-complexity with little performance loss. In addition, we develop an upper bound on the expected regret of this MAMAB based FL framework, which shows a linear growth over the logarithm of communication rounds, justifying its theoretical feasibility. Extensive experimental results are conducted to validate the effectiveness of our proposed algorithms, and the impacts of various parameters on the FL performance over wireless edge networks are also discussed. Copyright © 2021, The Authors. All rights reserved.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

The strength of structural diversity in online social networks

arXiv

引用

arXiv 2019年

作者： Zhang, Yafei Wang, Lin Zhu, Jonathan J.H. Wang, Xiaofan Pentland, Alex Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai 200240 China Department of Media and Communication School of Data Science City University of Hong Kong Hong Kong Hong Kong Department of Automation Shanghai University Shanghai 200444 China Media Lab Massachusetts Institute of Technology CambridgeMA02139 United States

Understanding the way individuals are interconnected in social networks is of prime significance to predict their collective outcomes. Leveraging a large-scale dataset from a knowledge-sharing website, this paper presents an exploratory investigation of the way to depict structural diversity in directed networks and how it can be utilized to predict one’s online social reputation. To capture the structural diversity of an individual, we first consider the number of weakly and strongly connected components in one’s contact neighborhood and further take the coexposure network of social neighbors into consideration. We show empirical evidence that the structural diversity of an individual is able to provide valuable insights to predict personal online social reputation, and the inclusion of coexposure network provides an additional ingredient to achieve that goal. After synthetically controlling several possible confounding factors through matching experiments, structural diversity still plays a nonnegligible role in the prediction of personal online social reputation. Our work constitutes one of the first attempts to empirically study structural diversity in directed networks and has practical implications for a range of domains, such as social influence and collective intelligence studies. Copyright © 2019, The Authors. All rights reserved.

关键词： Social networking (online)

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：