检索结果-内蒙古大学图书馆

international Journal of computers and Applications 2023年第5期45卷 379-390页

作者： Liu, J. Shao, H.J. Deng, X. Jiang, Y.T. School of Computer Jiangsu University of Science and Technology Zhenjiang China Department of Electrical and Computer Engineering University of Nevada Las VegasNV United States

The paper discusses the high computational costs associated with convolutional neural networks (CNNs) in real-world applications due to their complex structure, primarily in hidden layers. To overcome this issue, the paper proposes a novel channel pruning technique that leverages the correlation topology of feature maps generated by each CNNs layer to construct a network with fewer nodes, reducing computational costs significantly. Redundant channels exhibit a high degree of topological similarity and tend to increase as the number of network layers rises. Removing the channel corresponding to highly correlated feature maps allows retrieval of the ‘base’ set of characteristics needed by subsequent layers. The proposed channel pruning technique provides a promising approach to reducing the computational costs of deep convolutional neural networks while maintaining high performance levels. By designing a network structure optimized for specific input data types, the method results in more efficient and effective machine learning models. The pruning operation requires fine-tuning to optimize network performance, and experiments using X-ray, chest CT, and MNIST images show that the pruned network can eliminate approximately 80% of redundant channels with minimal performance deterioration (maintaining original CNNs performance at 99.2%). © 2023 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

DeRL: Coupling Decomposition in Action Space for Reinforcement Learning Task

引用

IEEE Transactions on Emerging Topics in Computational Intelligence 2024年第1期8卷 1030-1043页

作者： He, Ziming Li, Jingchen Wu, Fan Shi, Haobin Hwang, Kao-Shing Northwestern Polytechnical University School of Computer Science Xi'an710072 China National Sun Yat-sen University Department of Electrical Engineering Kaohsiung80424 Taiwan

This paper is concerned with complex reinforcement learning tasks whose observations are difficult to characterize as appropriate inputs for policy mapping. The representation learning technique is leveraged to extract the features from the observations for optimal action generation. In literature, the action vector, which consists of actions in each dimension, is usually learned from the full features. However, we find empirically that different actions may be only highly related to small part of the features and weakly depend on the rest. Therefore, this unified learning strategy may lead to performance degradation, and a separate learning method is motivated. In this paper, we propose a novel method called Decoupled Reinforcement Learning (DeRL) that decomposes action space by replacing the policy network with decoupled sub-policy group. To cater to all types of tasks where agents. actions in different dimensions can be either weakly correlated or strongly correlated, the Bidirectional Recurrent Neural Network (Bi-RNN) is added as an essential component to capture further shared features for generating more accurate action. In this framework, the decoupled policy network maintains joint representations required by the decision of all actions in different dimensions while decreasing preference in the learning process. In addition, we give a theoretical analysis of DeRL from the perspective of information theory, which shows the difference in information loss between DeRL and others. The performance of the proposed method has been verified by contrastive experiments on 12 tasks, including Mujoco, Atari, and other popular environments. © 2017 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Blockchain-based crowdsourcing for human intelligence tasks with dual fairness

引用

Blockchain(Research and Applications) 2024年第4期5卷 1-13页

作者： Yihuai Liang Yan Li Byeong-Seok Shin School of Information Science and Technology Southwest Jiaotong UniversityChengdu 611756China Department of Electrical and Computer Engineering Inha UniversityIncheon 22212Republic of Korea

Human intelligence tasks(HITs),such as labeling images for machine learning,are widely utilized for crowdsourcing human *** crowdsourcing platforms face challenges of a single point of failure and a lack of service *** blockchain-based crowdsourcing approaches overlook the low scalability problem of permissionless blockchains or inconveniently rely on existing ground-truth data as the root of trust in evaluating the quality of workers’*** propose a blockchain-based crowdsourcing scheme for ensuring dual fairness(i.e.,preventing false reporting and free riding)and improving on-chain efficiency concerning on-chain storage and smart contract *** proposed scheme does not rely on trusted authorities but rather depends on a public blockchain to guarantee dual *** efficient and publicly verifiable truth discovery scheme is designed based on majority voting and cryptographic *** truth discovery scheme aims at inferring ground truth from workers’*** ground truth is further utilized to estimate the quality of workers’***,a novel blockchain-based protocol is designed to further reduce on-chain costs while ensuring *** scheme has O(n)complexity for both on-chain storage and smart contract computation,regardless of the number of questions,where𝑛denotes the number of *** security analysis is provided,and extensive experiments are conducted to evaluate its effectiveness and performance.

关键词： Crowdsourcing Dual fairness Blockchain Human intelligence task Truth discovery

来源：评论

学校读者我要写书评

暂无评论

Optimal Design of a Planning and Contracting Framework to Enable Vehicle to Building and Grid Services via Demand Response

引用

IEEE Transactions on Transportation Electrification 2025年第3期11卷 7542-7556页

作者： Abd Elaziz Elsayed, Ahmed Saxena, Shivam Essa Zidan Farag, Hany York University Lassonde School of Engineering Electrical Engineering and Computer Science Department Keele Campus TorontoONM3J 1P3 Canada University of New Brunswick Department of Electrical and Computer Engineering FrederictonNBE3B5A3 Canada

Bidirectional electric vehicle (EV) charging enables stored energy to reduce peak loads for vehicle to buildings (V2Bs) and the vehicle to grid (V2G). However, building owners investing in V2B infrastructure while generating revenue from V2G services face challenges in planning and coordinating with EV owners due to uncertainties in their schedules and profit-sharing expectations. Additionally, misaligned building and grid peak times can create conflicts between V2B and V2G goals, which may negatively impact the building electricity bill. Unlike previous studies that used a contract-free approach for aggregating V2B and V2G, resulting in inconsistent participation, this article proposes a novel planning and contracting framework that enables building owner to determine the optimal contract parameters with EV owners. These parameters include minimum participation time in DR events, minimum arrival state of charge (SoC), and permitted emergency departure hours. The framework supports V2B aggregation with on-site distributed energy resources (DERs) for DR and V2G services, ensuring transparency and fairness through shared profits and performance-based penalties, while compensating building electricity bills due to V2G activities. The framework is composed of a tri-stage optimization process that uses Monte Carlo simulations to generate EV owner profit assessments, select optimal EV candidates based on charger availability, estimate contract parameters with profit/penalty sharing, and assign contracts between power system operators (PSOs), building owners, and EV owners under multiple virtual scenarios. Simulation results validate the contract parameter assessment and its performance via a 3-day case study with real-world datasets that demonstrates net revenue generation of {\} 209 for each EV owner and {\} 950 for building owners during DR events with a return on investment (ROI) of 134.5% and 130.7% for the EVs and Building owners, respectively. © 2015 IEEE.

关键词： Vehicle-to-grid

来源：评论

学校读者我要写书评

暂无评论

Wideband Slot Antenna Array with Stable Low Sidelobe Performance

引用

IEEE Antennas and Wireless Propagation Letters 2025年第5期24卷 1248-1252页

作者： Zhang, Fengyuan Zhang, Hanlin You, Qingchun Huang, Jifu Ningbo University Faculty of Electrical Engineering and Computer Science Zhejiang Ningbo315211 China University of Birmingham School of Engineering B15 2TT United Kingdom

This letter presents a dispersion compensation method that integrates a comb-shaped slow-wave structure with a defected ground structure (DGS) to achieve wideband low-sidelobe performance. This method effectively ensures phase balance within the feed network, achieving a stable low sidelobe level over a wide frequency range. A prototype planar array antenna operating in the 21.6-30.8 GHz range was designed, fabricated, and measured. The experimental results demonstrate that the antenna achieves a fractional bandwidth (FBW) of 35.1%, with the first sidelobe level maintained below -21 dB across the entire band and a fluctuation confined within ±1.3 dB. © 2002-2011 IEEE.

关键词： Defected ground structures

来源：评论

学校读者我要写书评

暂无评论

Learning Multiagent Options for Tabular Reinforcement Learning using Factor Graphs

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第5期4卷 1141-1153页

作者： Chen, Jiayu Chen, Jingdi Lan, Tian Aggarwal, Vaneet Purdue University School of Industrial Engineering West LafayetteIN47907 United States George Washington University Department of Electrical and Computer Engineering WashingtonDC20052 United States Purdue University School of Industrial Engineering The School of Electrical and Computer Engineering West LafayetteIN47907 United States King Abdullah University of Science and Technology Department of Computer Science Thuwal23955 Saudi Arabia

Covering option discovery has been developed to improve the exploration of reinforcement learning in single-agent scenarios, where only sparse reward signals are available. It aims to connect the most distant states identified through the Fiedler vector of the state transition graph. However, the approach cannot be directly extended to multiagent scenarios, since the joint state space grows exponentially with the number of agents, thus prohibiting efficient option computation. Existing research adopting options in multiagent scenarios still relies on single-agent algorithms and fails to directly discover joint options that can improve the connectivity of the joint state space. In this article, we propose a new algorithm to directly compute multiagent options with collaborative exploratory behaviors while still enjoying the ease of decomposition. Our key idea is to approximate the joint state space as the Kronecker product of individual agents' state spaces, based on which we can directly estimate the Fiedler vector of the joint state space using the Laplacian spectrum of individual agents' transition graphs. This decomposition enables us to efficiently construct multiagent joint options by encouraging agents to connect the subgoal joint states, which are corresponding to the minimum or maximum of the estimated joint Fiedler vector. Evaluation on multiagent collaborative tasks shows that our algorithm can successfully identify multiagent options and significantly outperforms prior works using single-agent options or no options, in terms of both faster exploration and higher cumulative rewards. © 2020 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Transient Stability Analysis of Low-Inertia Shipboard electrical Systems

Transient Stability Analysis of Low-Inertia Shipboard Electr...

引用

2024 IEEE international Conference on Modeling, Simulation and Intelligent Computing, MoSICom 2024

作者： Uddin, Imran Safavizadeh, Arash Tahim, Ekamjot Singh Ebrahimi, Seyyedmilad Jatskevich, Juri University of British Columbia Department of Electrical and Computer Engineering Vancouver Canada Washington State University School of Electrical Engineering and Computer Science Pullman United States

ISBN: (纸本)9798331533311

Transient stability analysis (TSA) of synchronous generators (SGs) is essential for reliable operation of electrical systems. Conventionally, the equal area criterion (EAC) is used to assess the transient stability of SGs, assuming that they are connected to a bus with a fixed frequency. However, during disturbances in low-inertia systems, e.g., small microgrids and shipboard power systems, the frequency of the main ac bus may noticeably deviate. This paper proposes a new formulation within the EAC method for calculating the critical clearing time (CCT) of SGs in low-inertia electrical systems, considering the effect of variations in system frequency during TSA. To verify the effectiveness of the extended EAC, the CCTs of SGs in a multi-machine shipboard power system are obtained using the proposed formulation and compared with time-domain simulation results. The proposed EAC method estimates the CCTs more accurately when the main ac bus frequency varies following a disturbance. © 2024 IEEE.

关键词： Transient stability

来源：评论

学校读者我要写书评

暂无评论

An artificial systems,computational experiments and parallel execution-based surface electromyogram-driven anti-disturbance zeroing neurodynamic strategy for upper limb human-robot interaction control

引用

CAAI Transactions on Intelligence Technology 2024年第2期9卷 511-525页

作者： Yongbai Liu Keping Liu Gang Wang Jiliang Zhang Yao Chou Zhongbo Sun Department of Mechanical Engineering Changchun University of TechnologyChangchunChina Department of Control Engineering Changchun University of TechnologyChangchunChina School of Electrical and Information Engineering Jilin Engineering Normal UniversityChangchunChina College of Information Science and Engineering Northeastern UniversityShenyangChina Department of Electrical and Computer Engineering Brigham Young UniversityProvoUtahUSA

In recent years,intelligent robots are extensively applied in the field of the industry and intelligent rehabilitation,wherein the human-robot interaction(HRI)control strategy is a momentous part that needs to be ***,the efficacy and robustness of the HRI control algorithm in the presence of unknown external disturbances deserve to be *** deal with these urgent issues,in this study,artificial systems,computational experiments and a parallel execution intelligent control framework are constructed for the HRI *** upper limb-robotic exoskeleton system is re-modelled as an artificial *** on surface electromyogram-based subject's active motion intention in the practical system,a non-convex function activated anti-disturbance zeroing neurodynamic(NC-ADZND)controller is devised in the artificial system for parallel interaction and HRI control with the practical ***,the linear activation function-based zeroing neurodynamic(LAF-ZND)controller and proportionalderivative(posterior deltoid(PD))controller are presented and *** results substantiate the global convergence and robustness of the proposed controller in the presence of different external *** addition,the simulation results verify that the NC-ADZND controller is better than the LAF-ZND and the PD controllers in respect of convergence order and anti-disturbance characteristics.

关键词： neural network pattern recognition robotics signal processing

来源：评论

学校读者我要写书评

暂无评论

Medical knowledge graph question answering for drug-drug interaction prediction based on multi-hop machine reading comprehension

引用

CAAI Transactions on Intelligence Technology 2024年第5期9卷 1217-1228页

作者： Peng Gao Feng Gao Jian-Cheng Ni Yu Wang Fei Wang Qiquan Zhang School of Cyber Science and Engineering Qufu Normal UniversityQufuChina School of Computer Science and Technology East China Normal UniversityShanghaiChina Network and Information Center Qufu Normal UniversityQufuChina School of Electronics and Information Engineering Harbin Institute of TechnologyShenzhenChina School of Electrical Engineering and Telecommunications University of New South WalesSydneyNSWAustralia

Drug-drug interaction(DDI)prediction is a crucial issue in molecular *** methods of observing drug-drug interactions through medical experiments require significant resources and *** authors present a Medical Knowledge Graph Question Answering(MedKGQA)model,dubbed MedKGQA,that predicts DDI by employing machine reading comprehension(MRC)from closed-domain literature and constructing a knowledge graph of“drug-protein”triplets from open-domain *** model vectorises the drug-protein target attributes in the graph using entity embeddings and establishes directed connections between drug and protein entities based on the metabolic interaction pathways of protein targets in the human *** aligns multiple external knowledge and applies it to learn the graph neural *** bells and whistles,the proposed model achieved a 4.5%improvement in terms of DDI prediction accuracy compared to previous state-of-the-art models on the QAngaroo MedHop *** results demonstrate the efficiency and effectiveness of the model and verify the feasibility of integrating external knowledge in MRC tasks.

关键词： artificial neural network data analysis document handling natural language processing

来源：评论

学校读者我要写书评

暂无评论

Low-complexity signal detection for massive MIMO systems via trace iterative method

引用

Journal of Systems engineering and Electronics 2024年第3期35卷 549-557页

作者： IMRAN A.Khoso ZHANG Xiaofei ABDUL Hayee Shaikh IHSAN A.Khoso ZAHEER Ahmed Dayo College of Electronic and Information Engineering Nanjing University of Aeronautics and AstronauticsNanjing 211106China School of Electrical Engineering Korea UniversitySeoul 02841Korea School of Mathematics South China University of TechnologyGuangzhou 510641China Department of Computer Science Huanggang Normal UniversityHuanggang 438000China

Linear minimum mean square error(MMSE)detection has been shown to achieve near-optimal performance for massive multiple-input multiple-output(MIMO)systems but inevitably involves complicated matrix inversion,which entails high *** avoid the exact matrix inversion,a considerable number of implicit and explicit approximate matrix inversion based detection methods is *** combining the advantages of both the explicit and the implicit matrix inversion,this paper introduces a new low-complexity signal detection ***,the relationship between implicit and explicit techniques is ***,an enhanced Newton iteration method is introduced to realize an approximate MMSE detection for massive MIMO uplink *** proposed improved Newton iteration significantly reduces the complexity of conventional Newton ***,its complexity is still high for higher ***,it is applied only for first two *** subsequent iterations,we propose a novel trace iterative method(TIM)based low-complexity algorithm,which has significantly lower complexity than higher Newton *** guarantees of the proposed detector are also *** simulations verify that the proposed detector exhibits significant performance enhancement over recently reported iterative detectors and achieves close-to-MMSE performance while retaining the low-complexity advantage for systems with hundreds of antennas.

关键词： signal detection low-complexity linear minimum mean square error(MMSE) massive multiple-input multiple-output(MIMO) trace iterative method(TIM)

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：