检索结果-内蒙古大学图书馆

3rd International Conference on Embedded Systems and Artificial Intelligence, ESAI 2024

作者： Sharan, Sudeep Thanh, Hung Do Nauth, Peter Dominguez-Jimenez, Juan Jose Frankfurt University of Applied Sciences Faculty of Computer Science and Engineering Frankfurt am Main Germany University of Cádiz UCASE Software Engineering Group School of Engineering Cádiz Spain

ISBN: (纸本)9798331522056

This paper presents an object detection and classification of the objects using Deep Learning (DL). The integration of object detection algorithms and depth camera developed is capable of providing robots, such as self developed assistive intelligent robot called 'ROSWITHA' (RObot System WITH Autonomy) with the ability to detect an object and interpret their surrounding environment. It also provides the real-world coordinates/position of the object in that environment, which helps the robot to perform some applications, such as grabbing and serving, etc. You Only Look Once (YOLO) is one of the most popular object detection algorithms which is treated as a single regression problem and is processed by a single neural network. In this work, the YOLOv5 (YOLO version 5) network is implemented on a depth camera 'RealSense' for processing object detection and estimating the position of the objects. The model is trained on reduced/filtered application-required objects (e.g. bottle and cup) from the available COCO dataset, and the training outcome alongside the performance of the overall system was evaluated. The results exhibited a satisfactory performance of the model in accurately detecting and classifying the objects with a high confidence score ranging from the least to7 2% to the highest of9 0%. The confidence score could also be adjusted by the user on the developed GUI to attain the most certain predictions. © 2024 IEEE.

关键词： Bottles

来源：评论

学校读者我要写书评

暂无评论

Masked Image Self-Learning and Knowledge Distillation for Source-Free Unsupervised Domain Adaptation in Crowd Counting 2

Masked Image Self-Learning and Knowledge Distillation for So...

引用

2nd International Conference on Image Processing, computer Vision and Machine Learning, ICICML 2023

作者： Zeng, Jia Ma, Chaoqun Shao, Penghui Qing, Anyong Wang, Yang Southwest Jiaotong University School of Electrical Engineering Chengdu China Concordia University Department of Computer Science and Software Engineering Montreal Canada

ISBN: (纸本)9798350331417

Unsupervised Domain Adaptation (UDA) techniques leverage labeled data from a source domain to adapt to unlabeled data from a target domain, offering a primary solution for addressing cross-domain challenges in crowd counting. However, in real-world scenarios, accessing source data during the model adaptation process within the target domain can be impractical due to privacy and security concerns. This gives rise to a practical yet challenging issue known as Source-Free Unsupervised Domain Adaptation (SFUDA), where model adaptation is limited to the utilization of an existing source model and unlabeled target data. To tackle this challenge, we propose an algorithm that combines mask image self-learning and knowledge distillation to address SFUDA in crowd counting effectively. Specifically, the student model uses masked images to self-learn at both feature and output levels, exploring clues and contextual relationships around masked image blocks to predict occluded areas. Furthermore, we employ knowledge distillation to boost the model's performance within the teacher-student framework, encompassing feature extraction and density map generation stages. To mitigate cumulative errors, we stochastically recovery a subset of parameters from the source model. We conducted extensive experiments to compare our approach with a modified SFUDA algorithm, and the results demonstrate competitive performance across multiple datasets. © 2023 IEEE.

关键词： Crowd counting Domain adaptation Knowledge distillation Self-learning Source-free

来源：评论

学校读者我要写书评

暂无评论

Semantic-Aware Pseudo-Labeling for Unsupervised Meta-Learning

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5475-5488页

作者： Ouyang, Tianran Dong, Xingping Ye, Mang Du, Bo Shao, Ling Shen, Jianbing Wuhan University School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan430072 China University of Chinese Academy of Sciences UCAS-Terminus AI Lab Beijing101408 China University of Macau State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science 314100 China

In unsupervised meta-learning, the clustering-based pseudo-labeling approach is an attractive framework, since it is model-agnostic, allowing it to synergize with supervised algorithms to learn from unlabeled data. However, the pseudo-labels suffer from clustering noise and semantic chaos problems, further impacting the effectiveness of meta-learning. In this paper, we analyze and optimize the pseudo-labeling process, including encoding and clustering, aiming to generate semantic-like pseudo-labels to narrow the gap between unsupervised and supervised meta-learning. First, during the encoding, we observe that the embedding space of existing methods lacks clustering-friendly properties, which is the primary reason for clustering noise. To address this issue, we minimize the inter-to-intra-class similarity ratio to generate clustering-friendly embedding features and validate our approach through comprehensive experiments. Then, during the clustering, we find that the semantic quality of pseudo-labels is not adequately controlled, resulting in semantic chaos of pseudo-labels. We propose a semantic-stability index to measure the semantic quality of pseudo-labels quantitatively. Based on this index, we propose the Semantic-aware Pseudo-label Reassignment mechanism to generate semantic-like pseudo-labels for all samples. Our approach is model-agnostic and can easily be integrated into existing supervised methods. To demonstrate its generalization ability, we integrate it into two representative algorithms: MAML and EP. The results on three main few-shot benchmarks clearly show that the proposed method achieves significant improvement compared to state-of-the-art models. Notably, our approach also outperforms the corresponding supervised method in three tasks. © 1979-2012 IEEE.

关键词： Latent semantic analysis

来源：评论

学校读者我要写书评

暂无评论

An application on forecasting for stock market prices: hybrid of some metaheuristic algorithms with multivariate adaptive regression splines

引用

International Journal of Intelligent Computing and Cybernetics 2023年第4期16卷 847-866页

作者： Dilek Sabancı Serhat Kılıçarslan Kemal Adem Department of Mathematics Tokat Gaziosmanpas¸a UniversityTokatTurkiye Department of Software Engineering Bandırma Onyedi Eylul UniversityBalıkesirTurkiye Department of Computer Engineering Sivas University of Science and TechnologySivasTurkiye

Purpose:Borsa Istanbul 100 Index,known as BIST100,is the main indicator to measure the performance of the 100 highest stocks publicly traded in Borsa Istanbul concerning market and trading *** 100 index prediction is a popular research domain for its complex data structure caused by stock price,commodity,interest rate and exchange rate *** study proposed hybrid models using both Genetic,Particle Swarm Optimization,Harmony Search and Greedy algorithms from metaheuristic algorithms approach for dimension reduction,and MARS for ***/methodology/approach:This paper aims to model in the simplest way through metaheuristic algorithms hybridized with the MARS model the effects of stock,commodity,interest and exchange rate variables on BIST 100 during the Covid-19 pandemic period(in the process of closing)between January 2020 and June ***:The most suitable hybrid model was chosen as PSO&MARS by calculating the RMSE,MSE,GCV,MAE,MAD,MAPE and R2 measurements of training,test and overall dataset to check every model’s *** results demonstrated that the proposed PSO&MARS hybrid modeling procedure gave results both as good as the MARS model and a simpler and non-complex model ***/value:Using metaheuristic algorithms as a supporting tool for variable selection can help to identify important independent variables and contribute to the establishment of more non-complex ***,test and overall dataset to check every model’s efficiency.

关键词： Dimension reduction Metaheuristic algorithms MARS Feature selection BIST 100

来源：评论

学校读者我要写书评

暂无评论

An Air Quality Oriented Convolutional and Interaction Network for Traffic Flow Prediction in VANET 16

An Air Quality Oriented Convolutional and Interaction Networ...

引用

16th International Conference on Wireless Communications and Signal Processing, WCSP 2024

作者： Cheng, Hanlin Li, Xuan Zhou, Tianqing Yao, Yu Wang, Kan Zhou, Momiao School of Information and software Engineering East China Jiaotong University Nanchang China School of Information and Communication Engineering Hainan University Haikou China School of Computer Science and Engineering Xi'an University of Technology Xi'an China School of Computer Science and Information Engineering Hefei University of Technology Hefei China

ISBN: (纸本)9798350390643

To accurately predict traffic flow, it is necessary to take into account not only the volume of road traffic, but also external urban factors. Particularly, the correlations between urban air quality and traffic flow is particularly strong, However, due to the complexity of air composition and its nonlinear characteristics in spatial and temporal sequences, it remains underexplored in current research efforts. In this paper, we propose an air quality oriented convolutional and interaction network for traffic flow prediction in Vehicular ad-hoc network, AQCInet, which captures feature information from both types of data using convolutional block, integrates this feature information with interaction networks, and effectively represents the long-term and short-term dependencies in the sequential data, ulti-mately outputting short-term traffic flow prediction results. The experimental results show that AQCInet can effectively model traffic flow sequences with complex time dynamics by aggregating rich features from multiple resolutions. Compared with existing model, our proposed model significantly improves RMSE, MSE, and MAE through jointly considerations of the relationship between air quality, various air component concentrations, and traffic flow. © 2024 IEEE.

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

A Multi-feature Co-distillation Algorithm Based on Human Skeletal Point Behavior Recognition 24

A Multi-feature Co-distillation Algorithm Based on Human Ske...

引用

2nd International Conference on Advances in Artificial Intelligence and Applications, AAIA 2024

作者： Pan, Xuefeng Tian, Zhiqiang School of Computer Wuhan Qingchuan University Wuhan China School of Information Wuhan Software Engineering Vocational College Wuhan Open University Wuhan China School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9798400712883

Depth model-based behavior estimation of human skeletal points is widely used in the field of behavior recognition. In order to improve the accuracy of behavior recognition, the complexity and computation of the designed network model are increasing, while the mobile platforms have limited arithmetic power and memory, and thus cannot be deployed. To solve the above problems, the AGCN network, a skeletal point-based behavior recognition model, is clipped to compress the network size so that it can support common mobile platforms;and a multi-feature co-distillation method is proposed to efficiently migrate the performance of the original network to the target network, so that the performance of the clipped lightweight network tends to be close to that of the original network. The model training, testing, visualization, and ablation experiments are completed on the NTU-RGB+D60 data set. The experimental results show that the algorithm after lightweighting (clipping network and multi-feature co-distillation), can significantly reduce the computational amount of behavior recognition. © 2024 Copyright held by the owner/author(s).

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Optimization Performance Analysis of Different Initialization Methods in Single-Objective and Multi-Objective Chaotic Evolution 13

Optimization Performance Analysis of Different Initializatio...

引用

Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024

作者： Zhang, Xiaoyi Pei, Yan Jiang, Huiyan Graduate School of Computer Science and Engineering University of Aizu Aizuwakamatsu965-8580 Japan University of Aizu Computer Science Division Aizuwakamatsu965-8580 Japan Northeastern University College of Software Shenyang110819 China

ISBN: (纸本)9798350373332

Chaotic Evolution (CE) is a significantly faster and more robust method for solving single-objective and multi-objective optimization problems. However, there are various factors that can impact the performance of CE, with population initialization being one of them. A well-executed initialization method can contribute to the discovery of better solutions and enhance both convergence rate and accuracy. This paper conducts a comparative experiment on CE algorithms using six different initialization methods. The performance of the CE algorithms utilizing these various initialization methods is analyzed and discussed. To assess convergence rate and accuracy, eight single-objective optimization benchmark functions and five multi-objective optimization benchmark functions are employed in our analysis. The evaluation results reveal significant disparities among the six initialization methods, indicating that different approaches to initialization have a substantial influence on the optimization performance of the CE algorithm. © 2024 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Cost-Effective Clustering by Aggregating Local Density Peaks 28th

Cost-Effective Clustering by Aggregating Local Density Peak...

引用

28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

作者： Xie, Wen-Bo Chen, Bin Shi, Jun-Hao Lee, Yan-Li Wang, Xin Fu, Xun School of Computer Science Southwest Petroleum University Chengdu610500 China School of Computer and Software Engineering Xihua University Chengdu610039 China

ISBN: (纸本)9783031306778

Hierarchical clustering algorithms that provide tree-shaped results can be regarded as data summarization and thus play an important role in the application of knowledge discovery and data mining. However, such structured result also brings a challenge, i.e., a difficult trade-off between complexity (time and space) and quality. To tackle of this issue, we propose a newly designed agglomerative algorithm for hierarchical clustering in this paper, which merges data points into tree-shaped sub-clusters via the operations of nearest-neighbor chain searching and determines the proxy of each sub-cluster by the process of local density peak detection. Extensive experimental studies on real-world and synthetic datasets show that our method performs well by outperforming other baselines in accuracy, response time, and memory footprint. Meanwhile, our method can scale to half a million data points on a personal computer, further verifying its cost-effectiveness. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Personal computers

来源：评论

学校读者我要写书评

暂无评论

Overbooking-Enabled Task Scheduling and Resource Allocation in Mobile Edge Computing Environments

引用

Intelligent Automation & Soft Computing 2023年第7期37卷 1-16页

作者： Jixun Gao Bingyi Hu Jialei Liu Huaichen Wang Quanzhen Huang Yuanyuan Zhao School of Computer Henan University of EngineeringZhengzhou451191China School of Computer Science and Technology Henan Polytechnic UniversityJiaozuo454000China School of Computer Hubei University of Arts and ScienceXiangyang441053China School of Software Engineering Anyang Normal UniversityAnyang455000China School of Computer and Information Engineering Henan Normal UniversityXinxiang453003China The Publicity Department Zhengzhou University of TechnologyZhengzhou450044China

Mobile Edge Computing(MEC)is proposed to solve the needs of Inter-net of Things(IoT)users for high resource utilization,high reliability and low latency of service ***,the backup virtual machine is idle when its primary virtual machine is running normally,which will waste *** the backup virtual machine under the above circumstances can effectively improve resource ***,these virtual machines are deployed into slots randomly,and then some tasks with cooperative relationship are off-loaded to virtual machines for *** deployment locations have different resource utilization and average service response *** want toﬁnd a balanced solution that minimizes the average service response time of the IoT application while maximizing resource *** this paper,we propose a task scheduler and exploit a Task Deployment Algorithm(TDA)to obtain an optimal virtual machine deployment ***,the simulation results show that the TDA can signiﬁcantly increase the resource utilization of the system,while redu-cing the average service response time of the application by comparing TDA with the other two classical *** experimental results conﬁrm that the perfor-mance of TDA is better than that of other two methods.

关键词： Mobile edge computing overbooking resource utilization service response time task deployment algorithm

来源：评论

学校读者我要写书评

暂无评论

Improving efficiency in rationale discovery for Out-of-Distribution molecular representations

Improving efficiency in rationale discovery for Out-of-Distr...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Zhang, Jiahui Li, Jiahe Du, Wenjie Zhang, Shuai Wu, Di Wang, Yang University of Science and Technology of China School of Software Engineering Suzhou China University of Science and Technology of China School of Computer Science and Technology Suzhou China

ISBN: (纸本)9798350337488

Graph Neural Networks (GNNs), as the dominant approach in Molecular Representation Learning (MRL), have exhibited remarkable efficacy in diverse tasks, including molecular property prediction and drug discovery. Considering the dynamic and diverse nature of test molecules in real-world contexts, the previous independently and identically distributed (i.i.d.) assumption for training and test molecules is not align with the requirements of practical applications in chemistry. Rationalization has been proposed to enhance the Out-of-Distribution (OOD) generalization of GNN, but a critical concern remains unexplored: low efficiency in rationale discovery. We consider this bottleneck is due to large search space and overly flexible modeling. Here, we introduce a framework called Molecule Stratifier and Invariant Rational Improver (MSIRI) to overcome the challenge. Specifically, MSIRI adopt vector quantization to obtain invariant rationales and the search space is narrowed by substructure and two assumptions based on subgraph matching. Experimental results on ten datasets demonstrate that our approach both achieves significant improvements across various GNN backbones and outperforms other six Out-of-Distribution method, clearly demonstrating its effectiveness in addressing the OOD challenges in MRL. © 2023 IEEE.

关键词： GNN invariant learning molecule representation learning Out-of-Distribution

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：