检索结果-内蒙古大学图书馆

2024 International Conference on Cloud and Network Computing, ICCNC 2024

作者： Chen, Yitian Liu, Bo Zhang, Shigeng Liu, Xuan Central South University School of Computer Science and Engineering Changsha China Hunan University College of Computer Science and Electronic Engineering Changsha China

ISBN: (纸本)9798350367805

In practical applications, implementing reinforcement learning in multi-Agent environments is very important. Most of the existing multi-Agent reinforcement learning algorithms lack communication capabilities, forcing agents to learn solely from their local observations. Some reinforcement learning algorithms that include communication also limit the number of communicating individuals. We propose a multi-Agent reinforcement learning algorithm that incorporates a local communication mechanism. This algorithm does not need to limit the number of individuals with whom the agent communicates. The agent can communicate with any number of agents in each time step, and the agent can choose whether to communicate and with whom to communicate. Besides, this algorithm requires no additional communication content and only uses the information obtained by the agent from the environment. Compared with recent methods, this method makes better use of the environmental information that the agent can obtain and plays a greater role in the team. This makes the actions of multi-Agents more in line with human group actions. To verify the effect of our algorithm, we conducted experiments on several mature open-source multi-Agent reinforcement learning environments. The experimental results show that our algorithm is better than MADDPG and M3DDPG. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Factors affecting forensic electric network frequency matching-A comprehensive study

引用

Digital Communications and Networks 2024年第4期10卷 1121-1130页

作者： Guang Hua Qingyi Wang Dengpan Ye Haijian Zhang Guoyin Wang Shuyin Xia School of Electronic Information Wuhan UniversityWuhan430072China Institute for Infocomm Research(I2R) Agency for ScienceTechnology and Research(A*STAR)Singapore138632Singapore School of Cyber Science and Engineering Wuhan UniversityWuhan430072China School of Computer Science and Technology Chongqing University of Posts and TelecommunicationsChongqing400065China

The power system frequency fluctuations could be captured by digital recordings and extracted to compare with a reference database for forensic timestamp *** is known as the Electric Network Frequency(ENF)criterion,enabled by the properties of random fluctuations and intra-grid *** essence,this is a task of matching a short random sequence within a long reference,whose accuracy is mainly concerned with whether this match could be uniquely *** this paper,we comprehensively analyze the factors affecting the reliability of ENF matching,including the length of test recording,length of reference,temporal resolution,and Signal-to-Noise Ratio(SNR).For synthetic analysis,we incorporate the first-order AutoRegressive(AR)ENF model and propose an efficient Time-Frequency Domain noisy ENF synthesis ***,the reliability analysis schemes for both synthetic and real-world data are respectively *** a comprehensive study,we quantitatively reveal that while the SNR is an important external factor to determine whether timestamp verification is viable,the length of test recording is the most important inherent factor,followed by the length of ***,the temporal resolution has little impact on ***,a practical workflow of the ENF-based audio timestamp verification system is proposed,incorporating the discovered results.

关键词： Digital forensics Audio forensics Data authentication Timestamp verification Electric network frequency criterion

来源：评论

学校读者我要写书评

暂无评论

Nanopore design of sulfur doped hollow carbon nanospheres for superior potassium-ion battery anodes

引用

Rare Metals 2024年第5期43卷 2103-2114页

作者： Zhen-Dong Liu Hui-Yan Feng Yu-Chen Wang Fei Wang Yue Liu Jian-Xiao Yang Yue Gu Jun Tan Chong Ye Cheng-Zhi Zhang Ji Hua Laboratory Foshan 528000China School of Chemical Engineering and Light Industry Guangdong University of TechnologyGuangzhou 510006China Hunan Province Key Laboratory for Advanced Carbon Materials and Applied Technology College of Materials Science and EngineeringHunan UniversityChangsha 410082China

Sulfur doped carbonaceous materials are promising anodes for potassium-ion batteries because of their ability to bridge active sites and induce C/S electron coupling,resulting in increased ion storage capacitance However,the large potassium ions could cause significant volume expansion and structure collapse during operation in sulfur doped carbonaceous anodes,which lead to rapidly capacity sacrifice during long-term *** design for anchoring sulfur atom in carbon skeleton is a novel way to alleviate the structure collapse and maintain the cycling ***,this study developed a controlled nanopore and sulfur doped carbon sphere structure(S-NPHCSs).In potassium-ion batteries S-NPHCSs anode demonstrated exceptional performance with a high reversible capacity of 247 mAh·g^(–1)after 50 cycles at 0.2 A·g^(–1)and delivered a long cycle stability of 600 cycles at a high current density of 1.0 A·g^(–1).Interconnected nanopores and doped sulfur structure not only expand the accumulation space and offer ample active sites for diffusion and adsorption of potassium ions,but also build stable channels through nanopore structure to ensure the cyclic *** finding provides a fundamental theory for designing nanopore structures and introducing sulfur doped carbonaceous materials to enhance capacitive potassium storage and long cycle stability.

关键词： Nanopore Carbon materials Sulfur doping Anode Potassium-ionbatteries

来源：评论

学校读者我要写书评

暂无评论

DHAA: Distributed heuristic action aware multi-agent path finding in high density scene

引用

Multimedia Tools and Applications 2024年 1-19页

作者： Zhou, Dongming Pang, Zhengbin Li, Wei School of Computer Science National University of Defense Technology Deya Road Hunan Changsha413000 China School of Computer and Electronic Information Guangxi University University Road Guangxi Nanning530004 China

Multi-agent path finding (MAPF) in highly structured environments is an exciting and complex problem. Compared with lower-density environments, the problems of agent credit allocation and differentiated communication need to be effectively solved in a grid world with complex structures. Through assistance and decentralized training between agents, which local agent information is shared through communication. Therefore, this paper proposes a multi-agent path finding model for distributed heuristic behavior awareness in high-density scenes(DHAA). First, this paper uses a query mechanism to assign different weights to the agents in the neighborhood based on priority order to distinguish the importance of different information to the agents. Then, according to the differentiated communication and collision detection strategies, selective communication is carried out with agents that can change their strategies to achieve conflict avoidance. Finally, we encoder the values of the observation and state networks and input them into the deep Q network to get the following action. By embedding collision-free heuristic search paths into graph networks, which the model ability to handle long tasks is enhanced. Experiments show that the performance of the DHAA model proposed in this paper is close to the performance of the coupled centralized algorithm. We demonstrate the effectiveness of the DHAA model in high-density scenarios by exploring different benchmarks and ablation experiments. Experiments show that deep Q networks can reduce the average step size and differentiated communication can reduce bandwidth. The DHAA model also shows robustness in untrained maps, improving 23 % over the baseline model. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Sub-terahertz transmissive reconfigurable intelligent surface for integrated beam steering and self-OOK-modulation

引用

Light(science & Applications) 2025年第1期14卷 131-141页

作者： Dongfang Shen Feng Lan Luyang Wang Tianyang Song Munan Yang Tianyu Hu Yueting Li Xiaolei Nie Jiayao Yang Shixiong Liang Hongxin Zeng Hui-Fang Zhang Pinaki Mazumder Ziqiang Yang Yaxin Zhang Tie Jun Cui Sichuan THz Communication Technology Engineering Research Center School of Electronic Science and EngineeringUniversity of Electronic Science and Technology of ChinaChengdu611731China Yangtze Delta Region Institute(Huzhou) University of Electronic Science and Technology of ChinaHuzhou313000China Zhangjiang Laboratory Shanghai201204China School of Microelectronics Tianjin UniversityTianjin300072China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of ChinaShenzhen518100China Department of Electrical Engineering and Computer Science University of MichiganAnn ArborMI48109USA Southeast University Nanjing210096China

Boasting superior flexibility in beam manipulation and a simpler framework than traditional phased arrays,terahertz metasurface-based phased arrays show great promise for 5G-A/6G communication *** with the reflective reconfigurable intelligent surface(reflective RIS),the transmissive RIS(TRIS)offers more feasibility for transceiver multiplexing systems to meet the growing demand for high-performance beam tracking in terahertz communication and radar ***,the terahertz TRIS encounters greater challenges in phase shift,beam efficiency,and complex ***,we propose a sub-terahertz TRIS based on the phase shift via Pancharatnam-Berry(PB)metasurface and self-on-off keying(OOK)modulation via Schottky *** electrically reconfigurable unit cell consists of a column-wise phase resonator and a rectangular *** experimental retrieved equivalent lumped-element circuit model is implemented in joint field-circuit simulations and is validated by experiments.A fabricated prototype demonstrates excellent performance of TRIS with the minimum insertion loss of 2.8 dB for operational states,large bandwidth nearly covering the entire W-band for 1-bit phase shift,deep OOK amplitude modulation of 12 dB,and wide scanning range of±60°with low specular *** further implement an integrated platform combining high-speed beam steering and spatial-light modulation,verifying the point-to-point signal transmissions in different directions using the TRIS *** proposed TRIS with high-performance and cost-effective fabrication makes it a promising solution to terahertz minimalist communication systems,radar,and satellite communication systems.

关键词： reconfigurable reflective resonator

来源：评论

学校读者我要写书评

暂无评论

Automatic emotion recognition using deep neural network

引用

Multimedia Tools and Applications 2025年 1-30页

作者： Sujatha, R. Chatterjee, Jyotir Moy Pathy, Baibhav Hu, Yu-Chen School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India Department of CSE Graphic Era University Dehradun India School of Electrical Engineering Vellore Institute of Technology Vellore India Department of Computer Science Tunghai University Taichung City Taiwan

Emotions are a vital semantic part of human correspondence. Emotions are significant for human correspondence as well as basic for human–computer cooperation. Viable correspondence between people is possibly achieved when both the importance and the emotion of the correspondence are perceived by all groups included. Understanding the significance of language has generally been concentrated on in natural language processing (NLP) as a semantic examination. In NLP, the text can be handled appropriately for classification. Emotion detection from facial emotion is the subfield of social signal processing applied in a wide assortment of regions, explicitly for human and PC collaboration. Many researchers have proposed various approaches, generally utilizing machine learning concepts. Automatic emotion recognition (AER) is significant for working with consistent intuitiveness between a person and a smart device toward fully acknowledging an intelligent society. Many researchers examined cross-lingual and multilingual speech emotion as a stage toward language-free emotion acknowledgment in natural speech. In the present work, we are proposing a deep learning-based AER system using four openly accessible datasets, namely Basic Arabic Vocal Emotions Dataset (BAVED), Acted Emotional Speech Dynamic Database (AESDD), Urdu written in Latin/Roman Script (URDU), and Toronto Emotional Speech Set (TESS), by utilizing the Jupyter notebook and a Python library for music and audio synthesis named Librosa. The experimental results exhibited that the proposed approach achieves better than the existing approaches, i.e., the accuracy of the proposed system with the URDU dataset is 96.24%, the TESS dataset is 99.10%, the AESDD dataset is 65.97%, and the BAVED dataset is 73.12%. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

High-areal-capacity/power lithium metal microbattery configuration based on the mechanically flexible,ultra-lightweight,nanocellulose framework

引用

Nano Research 2024年第9期17卷 8155-8162页

作者： Shaowen Li Ting Zhao Helin Wang Zhiqiao Wang Min Zhang Ahu Shao Jiacheng Liu Zhaohui Wang Yue Ma State Key Laboratory of Solidification Processing Centre for Nano Energy MaterialsSchool of Materials Science and EngineeringNorthwestern Polytechnical UniversityXi'an 710072China College of Materials Science and Engineering Hunan UniversityChangsha 410082China The Faculty of Printing Packaging Engineering and Digital Media TechnologyXi'an University of TechnologyXi'an 710048China

The ubiquitous implementation of integrated microelectronics requires the on-chip power sources featured with the lightweight configuration design,high areal-capacity-loadings as well as facile reaction kinetics that beyond the current available microbattery ***,this study constructs a mechanically flexible,nanocellulose fiber(NCF)reinforced microbattery configuration,which consists of metal-organic frameworks(ZIF-8)modified NCF as the separator(MOF@NCF),the carbonized MOF@NCF as the metallic deposition substrate(c-MOF@NCF)as well as gradient-structured LiFePO4 particles infiltrated in the NCF matrix(LFP@NCF)as the *** film-stacked,integrated NCF-based microbattery prototype not only achieves the facile reaction kinetics with homogenized,dendrite-free Li metal deposition at high-capacity-loadings(2 mAh·cm^(-2)),but also eliminates the necessary use of metallic current collector to maximize the electroactive mass ratio,which therefore enables the high energy density of 6.8 mWh·cm^(-2)at the power output of 1.36 mW·cm^(-2)as well as the robust cyclability upon various geometric flexing *** study presents a quantum leap towards the facile reaction kinetics and multi-scale interfacial stability for the flexible microbattery construction that based on the sustainable utilization of bio-scaffolds.

关键词： nanocellulose microbattery mechanical flexible ultra-lightweight high-areal-capacity

来源：评论

学校读者我要写书评

暂无评论

Multi-Level Feature Calibration for Adversarial Training Against Attack 10

Multi-Level Feature Calibration for Adversarial Training Aga...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Wang, Yancong Tao, Ze Zhang, Jian Central South University School of Computer Science and Engineering Hunan Changsha China

ISBN: (纸本)9798331520861

While excelling in various vision tasks, deep neural networks remain vulnerable to adversarial samples, which cause erroneous predictions with minor input perturbations. In this work, we reveal the differences in feature mapping between adversarial and natural examples and propose a Multi-Level Feature Calibration (MLFC) training method based on this insight. We introduce a fine-grained calibration module that utilizes gradient information to adaptively adjust feature representations in the latent space. By calibrating features at multiple levels, the model achieves a more stable and accurate representation of the input data. Additionally, the objective function not only maximizes the confidence of each sample's target label but also incorporates a central metric loss to promote feature distributions with high inter-class separation. This loss faithfully reflects the ground distance between samples through the angular distance measure, which further mitigates the differences in the sample feature distributions. Experimental results from various white-box and black-box attack scenarios demonstrate the effectiveness of our approach in enhancing model robustness and generalization ability. Specifically, MLFC improves the robustness of existing adversarial training methods under PGD attacks by 3. 3 % on CIFAR-10 and 3. 8 8 % on CIFAR-100. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Selective Learning for Sample-Efficient Training in Multi-Agent Sparse Reward Tasks 26

Selective Learning for Sample-Efficient Training in Multi-Ag...

引用

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Chen, Xinning Liu, Xuan Ba, Yanwen Zhang, Shigeng Ding, Bo Li, Kenli College of Computer Science and Electronic Engineering Hunan University China School of Computer Science and Engineering Central South University China School of Computer Science National University of Defense Technology China

ISBN: (纸本)9781643684369

Learning effective strategies in sparse reward tasks is one of the fundamental challenges in reinforcement learning. This becomes extremely difficult in multi-agent environments, as the concurrent learning of multiple agents induces the non-stationarity problem and sharply increased joint state space. Existing works have attempted to promote multi-agent cooperation through experience sharing. However, learning from a large collection of shared experiences is inefficient as there are only a few high-value states in sparse reward tasks, which may instead lead to the curse of dimensionality in large-scale multi-agent systems. This paper focuses on sparse-reward multi-agent cooperative tasks and proposes an effective experience-sharing method, Multi-Agent Selective Learning (MASL), to boost sample-efficient training by reusing valuable experiences from other agents. MASL adopts a retrogression-based selection method to identify high-value traces of agents from the team rewards, based on which some recall traces are generated and shared among agents to motivate effective exploration. Moreover, MASL selectively considers information from other agents to cope with the non-stationarity issue while enabling efficient training for large-scale agents. Experimental results show that MASL significantly improves sample efficiency compared with state-of-the-art MARL algorithms in cooperative tasks with sparse rewards. © 2023 The Authors.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Survey on Task Scheduling Optimization Strategy under Multi-Cloud Environment

引用

computer Modeling in engineering & sciences 2023年第6期135卷 1863-1900页

作者： Qiqi Zhang Shaojin Geng Xingjuan Cai School of Computer Science and Technology Taiyuan University of Science and TechnologyTaiyuan030024China College of Electronics and Information Engineering Tongji UniversityShanghai201804China

Cloud computing technology is favored by users because of its strong computing power and convenient *** the same time,scheduling performance has an extremely efficient impact on promoting carbon ***,scheduling research in the multi-cloud environment aims to address the challenges brought by business demands to cloud data centers during peak ***,the scheduling problem has promising application prospects under themulti-cloud *** paper points out that the currently studied scheduling problems in the multi-cloud environment mainly include independent task scheduling and workflow task scheduling based on the dependencies between *** paper reviews the concepts,types,objectives,advantages,challenges,and research status of task scheduling in the multi-cloud *** scheduling strategies proposed in the existing related references are analyzed,discussed,and summarized,including research motivation,optimization algorithm,and related ***,the research status of the two kinds of task scheduling is compared,and several future important research directions of multi-cloud task scheduling are proposed.

关键词： Cloud computing task scheduling workflow review multi-cloud environment

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：