检索结果-内蒙古大学图书馆

24th IEEE International Conference on High Performance Computing and Communications, 8th IEEE International Conference on Data Science and Systems, 20th IEEE International Conference on Smart City and 8th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Application, HPCC/DSS/SmartCity/DependSys 2022

作者： Deng, Miao Li, Peng Liu, Yu Zhu, Rongjing Zhang, Yujie School of Computer Science and Technology Wuhan University of Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System China

ISBN: (纸本)9798350319934

Due to the interconnection between devices in the optical transmission network (OTN), when one device fails and generates alarm information, a large number of other invalid alarms are generated in OTN. How to mining association rules in alarm data is an important problem. In this paper, we investigate the association rules mining problem based on a statistical method. First, we remove the chattering alarm in preprocess to reduce data redundancy. Second, we propose a statistics-based association rule mining algorithm. We use a sliding time window to scan the alarm data, and we exploit the matrix to record the corresponding results. According to the matrix, we calculate the support and confidence of association rules. Compared with other traditional association rule mining algorithms, the proposed algorithm does not need to set the minimum support artificially, and the obtained association rules have strong causality. Through extensive experiments on syn-thetic and real OTN alarm datasets, we verify the effectiveness of our proposed algorithm and the correctness of the obtained association rules. © 2022 IEEE.

关键词： Association rules

来源：评论

学校读者我要写书评

暂无评论

Fair Incentive Allocation for Secure Data Sharing in Blockchain Systems: A Shapley Value Based Mechanism 3rd

Fair Incentive Allocation for Secure Data Sharing in Block...

引用

3rd International Conference on Autonomous Unmanned Systems, ICAUS 2023

作者： Yang, Ming Feng, Honglin Wang, Xin Jiao, Xuguo Wu, Xiaoming Qin, Shuwei Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China School of Information and Control Engineering Qingdao University of Technology Qingdao China State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Zhejiang University Hangzhou China Department of Public Security of Shandong Province Shandong Intelligent Public Security Joint Innovation Center Jinan China

ISBN: (纸本)9789819710829

In the contemporary era of information technology, the exponential surge in data has rendered colossal potential value, which would be fully unlocked through data circulation and sharing. However, when the data contains sensitive information, special attention must be paid to the data privacy and security problem for data sharing. A feasible solution is to incorporate blockchain technology into the data sharing system, and the new challenge is how to provide fair rewards for data owners to motivate them to contribute data. To address this challenge, in this paper, we first design a blockchain-based data sharing system to enhance the trust and reliability of the data shared among participants. Then we propose an incentive allocation method based on the Shapley value, which comprehensively incorporates each participant’s contribution while ensuring fairness for the allocation. Finally, multiple numerical simulations are conducted to validate the effectiveness of the proposed mechanism. © Beijing HIWING Scientific and Technological information Institute 2024.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Duration-Aware Task Assignment for Heterogeneous Mobility Users in Crowdsensing 25

Duration-Aware Task Assignment for Heterogeneous Mobility Us...

引用

25th IEEE International Conference on computer Supported Cooperative Work in Design, CSCWD 2022

作者： Liu, Zhiyuan Li, Peng Huang, Weiyi Wang, Jiahu Nie, Lei Wuhan University of Science and Technology School of Computer Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing Real-Time Industrial System China

ISBN: (纸本)9781665405270

This paper investigates the duration-aware task assignment problem for heterogeneous mobility users in MCS (DTAH problem). There are two types of users in MCS, vehicle users and pedestrian users. The vehicle user gets to the task fast, but the task duration is short. Pedestrian user meets the task duration limit, but it takes a long time to reach the task. Given a set of tasks and two types of users, each user, and each task has a duration limit. The DTAH problem is how to assign tasks to maximize the utility of the system. We prove that the DTAH problem is NP-hard by reducing the weighted maximum set coverage problem to the DTAH problem. Then, we solve the problem from a pedestrian perspective and a vehicle perspective. From the pedestrian user perspective, we propose a pedestrian user task assignment (PTA) algorithm based on the Kuhn-Munkres algorithm. From the vehicle user perspective, we propose a greedy vehicle user task assignment (VTA) algorithm. We prove that the VTA algorithm can obtain the approximate ratio of 1-1/e to the optimal value. Finally, we design the heterogeneous user task assignment (HTA) algorithm based on these PTA and VTA algorithms. Extensive experiments have proved that our proposed HTA algorithm achieves more efficient performance than other comparison algorithms. © 2022 IEEE.

关键词： Combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

Task Priority Aware Incentive Mechanism with Reward Privacy-Preservation in Mobile Crowdsensing 25

Task Priority Aware Incentive Mechanism with Reward Privacy-...

引用

25th IEEE International Conference on computer Supported Cooperative Work in Design, CSCWD 2022

作者： Wang, Jiahu Li, Peng Huang, Weiyi Chen, Zeqiang Nie, Lei Wuhan University of Science and Technology School of Computer Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing Real-Time Industrial System China

ISBN: (纸本)9781665405270

In mobile crowdsensing, there are generally two types of tasks, popular tasks, and unpopular tasks. For popular tasks, many people can perform that task, and the budget is overallocated. For unpopular tasks, fewer or no one is willing to complete them. How to motivate users to complete different popularity tasks during their work time is a challenging problem. In this paper, we design a task priority-aware incentive mechanism to solve this problem. First, we use hierarchical clustering to classify tasks into different priorities by considering their budgets, deadlines, and density distribution. The higher the task priority, the higher the extra rewards and credits the participating users get. To motivate more users to perform unpopular tasks, we give high priority to unpopular tasks. Then, we propose a greedy algorithm that allows more users to do high-priority tasks. However, too many budget adjustments can cause most users to do unpopular tasks as users are obsessed with their income. To prevent users from all selecting high-priority tasks, we further propose a differential privacy-based algorithm to protect task priority and reduce users' attention to their income. This algorithm protects users' income and allows users to focus more on task characteristics, such as task distribution and task deadline. Through many experiments in the reality Roma dataset, we evaluate two proposed algorithms compared with other solutions. © 2022 IEEE.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Decoupling Dipole Antennas with Dielectric Resonator

Decoupling Dipole Antennas with Dielectric Resonator

引用

Antennas and Propagation & USNC/URSI National Radio Science Meeting

作者： Zhichao Jin Kai Lu Nan Yang Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems School of Electronics and Information Technology (School of Microelectronics) Sun Yat-sen University Guangzhou China

ISBN: (数字)9798350369908

ISBN: (纸本)9798350369915

A decoupling method using a dielectric resonator (DR) is investigated for 1 × 2 multiple-input and multiple-output (MIMO) dipole antenna. The decoupling DR is placed between two dipole antennas separated by a half wavelength. It serves as a parasitic element, creating a neutralization path between the two antenna ports. By adjusting the DR size, the isolation level can be improved obviously. A dual-port antenna system is simulated. An overlapping bandwidth ( $\vert \mathrm{S}_{11}\vert$ < −10 dB & $\vert \mathrm{S}_{21}\vert$ < −20 dB) of 9.2% (3.36-3.61 GHz) is obtained. This design method may find applications in base station antenna arrays.

关键词： Couplings Mutual coupling Base stations Dipole antennas Design methodology Conferences Scattering

来源：评论

学校读者我要写书评

暂无评论

Label Noise Correction via Fuzzy Learning Machine 39

Label Noise Correction via Fuzzy Learning Machine

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Liang, Jiye Li, Yixiao Cui, Junbiao Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education School of Computer and Information Technology Shanxi University Shanxi Taiyuan 030006 China

ISBN: (纸本)157735897X

The ubiquitous and unavoidable label noise brings great challenges to the generalization performance of learning methods. Label noise correction aims to detect and correct label noise in the data, which is one of the most potential methods to address this challenge. Current methods for label noise filtering that utilize primitive features primarily concentrate on identifying noise, which often limits their capacity to adaptively learn features crucial for specific tasks, thereby resulting in a higher rate of noise identification within the noise recognition process. On the other hand, deep neural networks, endowed with robust feature extraction capabilities, typically exhibit lower noise identification, as they are prone to fitting noise patterns during the recognition process, potentially undermining their overall efficacy. Moreover, Fuzzy Learning Machine (FLM) excels not only in feature extraction but also in noise tolerance, adeptly navigating data uncertainties. FLM enhances the accuracy of the labels by calculating the membership degrees of samples across categories and determining their fuzzy memberships. The introduction of a two-stage FLM-based framework, which employs a secondary learning mechanism for precise noise filtering and correction, has shown substantial improvements in noise correction across various large-scale noisy datasets, thereby significantly enhancing samples' quality and boosting the generalization capabilities of classifiers. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Time Series Data Association Mining With High-Frequency Data Interference 24

Time Series Data Association Mining With High-Frequency Data...

引用

作者： Zhu, Rongjing Li, Peng Liu, Yu Deng, Miao School of Computer Science and Technology Wuhan University of Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System China

ISBN: (纸本)9798350319934

Time series data mining is an important problem in data mining. It discovers the intrinsic correlation rules by analyzing a large amount of time series data. However, high-frequency data may exist in the dataset. High-frequency data is the data that occurs frequently in time series data, which will affect the mining of association rules. Traditional time series association mining tends to classify high-frequency data as correlated data because they co-occur too many times. In this paper, we propose the list mining framework to mine the association data by list co-occurrence to achieve the mining of massive data. We divide it into three phases, list generation phase, matrix generation phase, and rule generation phase. First, we create data lists based on time windows. Second, we create the initial count matrix, update the count matrix based on the list, and propose a GCPM algorithm to generate a conditional probability matrix based on the count matrix. Third, we propose a List Mining (LM) algorithm to generate the initial rules based on the conditional probability matrix and optimize the initial rules to get the final association rules. In addition, we weigh the rules according to the list mining method and propose a Weighted List Mining (WLM) algorithm to filter rules by reducing the weight of frequent item interference. The experimental results show that LM and WLM can effectively improve the accuracy of mining results. © 2022 IEEE.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Ridge Gap Waveguide Fed Slot Array Antenna with Low Sidelobe Level

Ridge Gap Waveguide Fed Slot Array Antenna with Low Sidelobe...

引用

Cross Strait Quad-Regional Radio Science and Wireless technology Conference (CSRWTC)

作者： Shi-Kang Xie Nan Yang Kai Lu Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems School of Electronics and Information Technology (School of Microelectronics) Sun Yat-sen University Guangzhou China

ISBN: (数字)9798331507794

ISBN: (纸本)9798331507800

A low sidelobe slot array antenna based on ridge gap waveguide (RGW) is presented in this paper. Fed by two-layer RGW structure centrally, the slot array features symmetric field distribution. Offsetting the slots can fine tune the amplitude to obtain a quasi-Taylor distribution, thus resulting in a low sidelobe level. To verify the design idea, an 8-element slot array antenna was designed and simulated. Its simulated impedance bandwidth is found to be 8.9%, with a sidelobe level lower than -20 dB from 76 to 81 GHz.

关键词： Wireless communication Wireless sensor networks Slot antennas Gap waveguide Bandwidth Radar antennas Impedance Gain Tuning Antenna arrays

来源：评论

学校读者我要写书评

暂无评论

Embedding expert demonstrations into clustering buffer for effective deep reinforcement learning

引用

Frontiers of information technology & Electronic Engineering 2023年第11期24卷 1541-1556页

作者： Shihmin WANG Binqi ZHAO Zhengfeng ZHANG Junping ZHANG Jian PU Shanghai Key Laboratory of Intelligent Information Processing School of Computer ScienceFudan UniversityShanghai 200433China Institute of Science and Technology for Brain-inspired Intelligence Fudan UniversityShanghai 200433China

As one of the most fundamental topics in reinforcement learning(RL),sample efficiency is essential to the deployment of deep RL *** most existing exploration methods that sample an action from different types of posterior distributions,we focus on the policy sampling process and propose an efficient selective sampling approach to improve sample efficiency by modeling the internal hierarchy of the ***,we first employ clustering methods in the policy sampling process to generate an action candidate *** we introduce a clustering buffer for modeling the internal hierarchy,which consists of on-policy data,off-policy data,and expert data to evaluate actions from the clusters in the action candidate set in the exploration *** this way,our approach is able to take advantage of the supervision information in the expert demonstration *** on six different continuous locomotion environments demonstrate superior reinforcement learning performance and faster convergence of selective *** particular,on the LGSVL task,our method can reduce the number of convergence steps by 46.7%and the convergence time by 28.5%.Furthermore,our code is open-source for *** code is available at https://***/Shihwin/SelectiveSampling.

关键词： Reinforcement learning Sample eficiency Sampling process Clustering methods Autonomous driving

来源：评论

学校读者我要写书评

暂无评论

A Defense Framework for Backdoor Attacks in Federated Learning Based on Client-Server Detection 4

A Defense Framework for Backdoor Attacks in Federated Learni...

引用

4th International Conference on Industrial Automation, Robotics and Control Engineering, IARCE 2024

作者： Chen, Ding Li, Tianrui Yu, Cai Al-Huthaifi, Rasha Huang, Wei School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China School of Cybersecurity Chengdu University of Information Technology Sichuan Provincial Key Laboratory of Cyberspace Security Chengdu610225 China School of Cybersecurity Chengdu University of Information Technology Chengdu610225 China College of Computer and Data Science Fuzhou University Fuzhou350108 China

ISBN: (纸本)9798350380323

Addressing the challenge of backdoor attacks in federated learning (FL), various defense schemes are examined, resulting in the proposal of a comprehensive strategy that incorporates both client-side and server-side measures. This defense strategy is based on Dropout and LOF (Local Outlier Factor) outlier detection techniques. The model's resilience against data poisoning is enhanced by applying the Dropout technique on the client side. At the same time, the LOF outlier detection algorithm is utilized on the server side to identify potential abnormal model updates. Experimental results highlight the effectiveness of the proposed strategy, demonstrating a significant reduction in the Attack Success Rate (ASR) from 88.11% to 1.62% on the FMNIST dataset, and from 95.45% to only 0.25% on the EMNIST dataset. The success rate of backdoor attacks is notably diminished, confirming the strategy's strong backdoor defense capabilities. © 2024 IEEE.

关键词： Federated learning

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：