检索结果-内蒙古大学图书馆

8th International Conference on Electrical, Mechanical and computer engineering, ICEMCE 2024

作者： Zhao, Yuan Yuan, Shuangshuang Northeastern University at Qinhuangdao School of Computer and Communication Engineering Qinhuangdao China Northeastern University at Qinhuangdao Hebei Key Laboratory of Marine Perception Network and Data Processing Qinhuangdao China

ISBN: (纸本)9798331506230

In cognitive radio network, considering that secondary user (SU) packets waiting in the cache will stop waiting and exit the system because of impatience, a dynamic stop-adjustable control mechanism is proposed in a single-channel network environment where impatient SU packets leave the system in advance. The SU packets in the cache are divided into retry SU packets and newly arrived SU packets, assuming that only the newly arrived SU packets can stop and leave the system because of impatience. A discrete-time queueing model is established, a three-dimensional Markov chain is constructed, and SU packets' performance indicators are derived. Numerical experiments are carried out to verify the validity of the presented stop-adjustable control strategy. By comparing with the stopadjustable control mechanism for unclassified SU packets, it is found that the proposed stop-adjustable control mechanism for classified SU packets can effectively improve SU packets' performance. © 2024 IEEE.

关键词： Cognitive radio

来源：评论

学校读者我要写书评

暂无评论

Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts 41

Long-Tail Learning with Foundation Model: Heavy Fine-Tuning ...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Shi, Jiang-Xin Wei, Tong Zhou, Zhi Shao, Jie-Jing Han, Xin-Yan Li, Yu-Feng National Key Laboratory for Novel Software Technology Nanjing University China School of Artificial Intelligence Nanjing University China School of Computer Science and Engineering Southeast University China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education China

The fine-tuning paradigm in addressing long-tail learning tasks has sparked significant interest since the emergence of foundation models. Nonetheless, how fine-tuning impacts performance in long-tail learning was not explicitly quantified. In this paper, we disclose that heavy fine-tuning may even lead to non-negligible performance deterioration on tail classes, and lightweight fine-tuning is more effective. The reason is attributed to inconsistent class conditions caused by heavy fine-tuning. With the observation above, we develop a low-complexity and accurate long-tail learning algorithms LIFT with the goal of facilitating fast prediction and compact models by adaptive lightweight fine-tuning. Experiments clearly verify that both the training time and the learned parameters are significantly reduced with more accurate predictive performance compared with state-of-the-art approaches. The implementation code is available at https://***/shijxcs/LIFT. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Improving Multimodal Rumor Detection via Dynamic Graph Modeling 27th

Improving Multimodal Rumor Detection via Dynamic Graph Model...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Wu, Xinyu Hu, Xiaoxu Qin, Xugong Zhang, Peng Zeng, Gangyan Guo, Yu Zhao, Runbo Huang, Xinjian School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing China National Computer Network Response Technical Team/Coordination Center Technology Beijing China Laboratory for Advanced Computing and Intelligence Engineering Wuxi China

ISBN: (纸本)9783031784552

Due to the proliferation of rumors in social networks, automatic rumor detection has evoked increasing attention in recent years. Despite great progress achieved by exploiting multimodal features, existing works suffer from false discrimination issues due to insufficient multimodal modeling, mainly from two aspects: 1) neglect of the dynamicity of social networks. 2) misaligned multimodal features. To alleviate the issues, we propose DGM, Dynamic Graph Modeling for rumor detection. Firstly, dynamic graph attention is devised to exploit message propagation’s structural and temporal features. Secondly, we propose a modality-shared adapter to learn better multimodal representation. Thirdly, well-aligned visual-textual features are introduced to achieve better multimodality alignment and fusion, together with cross-modal attention and alignment supervision. We conduct extensive experiments on two public datasets, demonstrating the effectiveness and superiority of DGM. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Large-scale Satellite Optical network Simulation Architecture and Performance Evaluation

Large-scale Satellite Optical Network Simulation Architectur...

引用

2024 Asia Communications and Photonics Conference, ACP 2024 and International Conference on Information Photonics and Optical Communications, IPOC 2024

作者： Wang, Xinjun Zhang, Zhenrong Zhu, Kangqi Hua, Nan School of Computer Electronic and Information Guangxi Key Laboratory of Multimedia Communications and Network Technology Guangxi University Nanning China Department of Electronic Engineering Tsinghua University Beijing China

We evaluate the accuracy and performance of different simulation modes in MegaStarSim, analyze end-to-end routing strategies in large-scale satellite optical networks, and assess link quality across various inter-sate... 详细信息

ISBN: (纸本)9798350379266

关键词： Satellite links

来源：评论

学校读者我要写书评

暂无评论

Competitive Swarm Optimizer Based on Individual Learning Mechanism 5

Competitive Swarm Optimizer Based on Individual Learning Mec...

引用

5th International Conference on High Performance Big Data and Intelligent Systems, HDIS 2023

作者： Tang, JingQi Li, Wei Wang, Lei School of Computer Science and Engineering Xi’an University of Technology Xi’an China Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an China

ISBN: (纸本)9798350394160

In view of the disadvantages of the traditional competitive swarm optimizer (CSO), such as falling into local minimization or poor convergence accuracy, this paper proposed an enhanced CSO algorithm called competitive swarm optimizer based on individual learning mechanism (ILCSO). Firstly, the individual selection rate and the learning rate are designed to dynamically select the winner and loser. The losers are updated by the precise competitive learning strategy to improve the algorithm exploitation ability. Secondly, the mutation strategy based on performance improvement is introduced, which improves the search ability of algorithm. The proposed algorithm effectively balances the local search and global exploration with the individual learning mechanism, and improves the probability of finding the global optimal solution. Finally, ILCSO is compared with six classical meta-heuristic algorithms on CEC2014 benchmark functions. The Wilcoxon rank-sum test is used to demonstrate that the ILCSO is effective. Experimental results and statistical analysis show that ILCSO has higher convergence speed and convergence accuracy. ©2023 IEEE.

关键词： Swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

A Positive Influence Maximization Algorithm in Signed Social networks

引用

computers, Materials & Continua 2023年第8期76卷 1977-1994页

作者： Wenlong Zhu Yang Huang Shuangshuang Yang Yu Miao Chongyuan Peng College of Computer and Control Engineering Qiqihar UniversityQiqihar161006China Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis Qiqihar UniversityQiqihar161006China College of Teacher Education Qiqihar UniversityQiqihar161006China

The influence maximization(IM)problem aims to find a set of seed nodes that maximizes the spread of their influence in a social *** positive influence maximization(PIM)problem is an extension of the IM problem,which consider the polar relation of nodes in signed social networks so that the positive influence of seeds can be the most widely *** solve the PIM problem,this paper proposes the polar and decay related independent cascade(IC-PD)model to simulate the influence propagation of nodes and the decay of information during the influence propagation in signed social *** overcome the low efficiency of the greedy based algorithm,this paper defines the polar reverse reachable(PRR)set and devises a signed reverse influence sampling(SRIS)*** algorithm utilizes the ICPD model as well as the PRR set to select *** are two phases in *** is the sampling phase,which utilizes the IC-PD model to generate the PRR set and a binary search algorithm to calculate the number of needed PRR *** other is the node selection phase,which uses a greedy coverage algorithm to select optimal ***,Experiments on three real-world polar social network datasets demonstrate that SRIS outperforms the baseline algorithms in *** on the Slashdot dataset,SRIS achieves 24.7% higher performance than the best-performing compared algorithm under the weighted cascade model when the seed set size is 25.

关键词： Positive influencemaximization polar relation information decay reverse influence sampling

来源：评论

学校读者我要写书评

暂无评论

Fast Crop Pest Detection Using Lightweight Feature Extraction and Knowledge Distillation

Fast Crop Pest Detection Using Lightweight Feature Extractio...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Yang, Ze Jiang, Xianliang Jin, Guang Huang, Junkai Bai, Jie Yu, Dingxin Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo China Zhejiang Key Laboratory of Mobile Network Application Technology Ningbo China

ISBN: (纸本)9798350337020

Pest detection is critical for achieving effective pest control. However, the current deep learning-based pest detection algorithm is unsuitable for deployment on resource-limited edge devices due to its extensive computation and long inference time. Although lightweight models have been widely used for practical detection, their insufficient feature extraction capability leads to a decline in detection accuracy. This paper proposes a fast algorithm for crop pest detection based on lightweight feature extraction and knowledge distillation. Firstly, we introduce partial convolution and propose a lightweight feature extraction module, C3Faster, which reduces the model's computation and speeds up model inference while ensuring effective feature extraction. Secondly, we use knowledge distillation to improve the model's detection accuracy by using teacher networks to assist in training. Finally, we created a dataset, CropPest6, consisting of six crop pest categories and conducted experiments. The experimental results demonstrate that our method reduces the detection time, number of parameters, and computation by 17%, 38%, and 44%, respectively, compared to the baseline model. Furthermore, our method achieves 93.9% Precision, 93.6% Recall, and 97.5% mean Average Precision (mAP), demonstrating its practical suitability for fast crop pest detection. © 2023 IEEE.

关键词： crop pests deep learning object detection pest detection

来源：评论

学校读者我要写书评

暂无评论

A Trusted Privacy-Preserving Model for Cross-Chain Transactions Based on zk_SNARKs

A Trusted Privacy-Preserving Model for Cross-Chain Transacti...

引用

2023 International Conference on networking and network Applications, NaNA 2023

作者： Wang, Yichuan Tie, Jianhuan Hei, Xinhong Zhao, Li Zhang, Xiaohui School of Computer Science and Engineering Xi'an University of Technology Xi'an China Shaanxi Key Laboratory for Network Computing and Security Technolog Xi'an China

ISBN: (纸本)9798350327380

Currently, the privacy protection technology of blockchain is not mature enough, such as user data leakage and lack of anonymity exist, and these problems are especially serious when conducting cross-chain transactions. In this paper, we propose a method of cross-chain transactions based on zk_SNARKs in a trusted environment for cross-chain transactions for privacy protection. Cross-chain transactions are performed through the features of zk_SNARKs such as efficiency, privacy, and verifiability. zk_SNARKs needs to generate trusted settings with the help of a third party during the transaction, and the process has serious privacy issues, and the private parameters generated during the transaction may be maliciously attacked and thus obtained by attackers. To address this problem, we use a hardware technology SGX for trusted computing to encrypt and store some private parameters generated by zk_SNARKs, thus ensuring the security of cross-chain transactions. The experimental results show that the privacy protection scheme using SGX for cross-chain transactions takes less time than the traditional transaction privacy protection, with a verification time of 1.2s for a number of 10 sidechains, and is more efficient than the traditional cross-chain transaction privacy protection. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

A numerical magnitude aware multi-channel hierarchical encoding network for math word problem solving

引用

Neural Computing and Applications 2025年第3期37卷 1651-1672页

作者： Xu, Junjie Chen, Yuzhong Xiao, Lingsheng Liao, Hongmiao Zhong, Jiayuan Dong, Chen College of Computer and Data Science Fuzhou University Fujian Province Fuzhou350108 China Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fujian Province Fuzhou350108 China

Math word problem (MWP) represents a critical research area within reading comprehension, where accurate comprehension of math problem text is crucial for generating math expressions. However, current approaches still grapple with unresolved challenges in grasping the sensitivity of math problem text and delineating distinct roles across various clause types, and enhancing numerical representation. To address these challenges, this paper proposes a Numerical Magnitude Aware Multi-Channel Hierarchical Encoding network (NMA-MHEA) for math expression generation. Firstly, NMA-MHEA implements a multi-channel hierarchical context encoding module to learn context representations at three different channels: intra-clause channel, inter-clause channel, and context-question interaction channel. NMA-MHEA constructs hierarchical constituent-dependency graphs for different levels of sentences and employs a Hierarchical Graph Attention Neural network (HGAT) to learn syntactic and semantic information within these graphs at the intra-clause and inter-clause channels. NMA-MHEA then refines context clauses using question information at the context-question interaction channel. Secondly, NMA-MHEA designs a number encoding module to enhance the relative magnitude information among numerical values and type information of numerical values. Experimental results on two public benchmark datasets demonstrate that NMA-MHEA outperforms other state-of-the-art models. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Maximizing Influence in Temporal Social networks:A Node Feature-Aware Voting Algorithm

引用

computers, Materials & Continua 2023年第12期77卷 3095-3117页

作者： Wenlong Zhu Yu Miao Shuangshuang Yang Zuozheng Lian Lianhe Cui College of Computer and Control Engineering Qiqihar UniversityQiqihar161006China Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis Qiqihar UniversityQiqihar161006China College of Teacher Education Qiqihar UniversityQiqihar161006China

Influence Maximization(IM)aims to select a seed set of size k in a social network so that information can be spread most widely under a specific information propagation model through this set of ***,most existing studies on the IM problem focus on static social network features,while neglecting the features of temporal social *** bridge this gap,we focus on node features reflected by their historical interaction behavior in temporal social networks,i.e.,interaction attributes and self-similarity,and incorporate them into the influence maximization algorithm and information propagation ***,we propose a node feature-aware voting algorithm,called ISVoteRank,for seed nodes ***,before voting,the algorithm sets the initial voting ability of nodes in a personalized manner by combining their *** the voting process,voting weights are set based on the interaction strength between nodes,allowing nodes to vote at different extents and subsequently weakening their voting ability *** process concludes by selecting the top k nodes with the highest voting scores as seeds,avoiding the inefficiency of iterative seed selection in traditional voting-based ***,we extend the Independent Cascade(IC)model and propose the Dynamic Independent Cascade(DIC)model,which aims to capture the dynamic features in the information propagation process by combining node ***,experiments demonstrate that the ISVoteRank algorithm has been improved in both effectiveness and efficiency compared to baseline methods,and the influence spread through the DIC model is improved compared to the IC model.

关键词： Temporal social networks influence maximization voting strategy interactive properties self-similarity

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：