检索结果-内蒙古大学图书馆

Small Object Detection via Precise Region-Based Fully Convolutional Networks

computers, Materials & Continua 2021年第11期69卷 1503-1517页

作者： Dengyong Zhang Jiawei Hu Feng Li Xiangling Ding Arun Kumar Sangaiah Victor SSheng Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha University of Science and TechnologyChangsha410114China School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha410114China School of Computer Science and Engineering Hunan University of Science and TechnologyXiangtan411004China School of Computing Science and Engineering Vellore Institute of Technology(VIT)Vellore632014India Department of Computer Science Texas Tech UniversityLubbock79409TXUSA

In the past several years,remarkable achievements have been made in the field of object *** performance is generally improving,the accuracy of small object detection remains low compared with that of large object *** addition,localization misalignment issues are common for small objects,as seen in GoogLeNets and residual networks(ResNets).To address this problem,we propose an improved region-based fully convolutional network(R-FCN).The presented technique improves detection accuracy and eliminates localization misalignment by replacing positionsensitive region of interest(PS-RoI)pooling with position-sensitive precise region of interest(PS-Pr-RoI)pooling,which avoids coordinate quantization and directly calculates two-order integrals for position-sensitive score maps,thus preventing a loss of spatial precision.A validation experiment was conducted in which the Microsoft common objects in context(MS COCO)training dataset was *** showed an accuracy improvement of 3.7%for object detection tasks and an increase of 6.0%for small objects.

关键词： Small object detection precise R-FCN PS-Pr-RoI pooling two-stage detector

来源：评论

学校读者我要写书评

暂无评论

ExpertDRL: Request Dispatching and Instance Configuration for Serverless Edge Inference with Foundation Models

引用

IEEE Transactions on Mobile Computing 2025年

作者： Zeng, Yue Zhou, Junlong Ye, Baoliu Qu, Zhihao Guo, Song Gong, Tianjian Li, Pan Nanjing University of Science and Technology Department of Computer Science and Engineering Jiangsu Nanjing210094 China Nanjing University State Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing210023 China Hohai University Key Laboratory of Water Big Data Technology of Ministry of Water Resources Nanjing211100 China Hong Kong University of Science and Technology Hong Kong Southwest University College of Electronic and Information Engineering Chongqing400715 China

The prevalence of the pre-training & fine-tuning paradigm enables machine learning models to quickly adapt to various downstream tasks by fine-tuning pre-trained foundation models (FMs), greatly facilitating various IoT applications that rely on model inference in dynamic edge serverless environments. Efficiently dispatching inference requests and configuring instances to batch inference requests can significantly enhance resource efficiency. However, existing serverless inference solutions are tailored for traditional models, make coarse-grained request dispatching and instance configuration decisions, fail to exploit the shared model backbone characteristics of the FM and capture delayed rewards in dynamic environments, and ignore communication latency between edge sites, resulting in high costs and constraint violations. In this paper, we leverage our insight that fine-grained batch inference requests can effectively exploit the shared model backbone feature of FM to save monetary costs. We propose an algorithm that incorporates deep reinforcement learning (DRL) and expert intervention for fine-grained request dispatching and instance configuration, where the DRL component outputs fractional solutions as guidance, while the expert intervention module integrates our insights—batching reduces monetary costs at the expense of increased inference latency, whereas higher configurations shorten inference latency. This module rounds fractional solutions and adjusts instance configurations to search for optimal solutions while satisfying constraints, with theoretical guarantees rigorously proved. Finally, we conducted our experiments on an OpenFaas-based platform and simulator, and extensive trace-driven evaluation results show that ExpertDRL can save costs by up to 85.14% and improve request acceptance ratio by up to 26.93%, compared to the state-of-the-art solution. © 2002-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Prompt Learning on Temporal Interaction Graphs

arXiv

引用

arXiv 2024年

作者： Chen, Xi Zhang, Siwei Xiong, Yun Wu, Xixi Zhang, Jiawei Sun, Xiangguo Zhang, Yao Zhao, Feng Kang, Yulin Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China IFM Lab Department of Computer Science University of California DavisCA United States Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong Ant Group Shanghai Hangzhou China

Temporal Interaction Graphs (TIGs) are widely utilized to represent real-world systems, such as e-commerce and social networks. To facilitate representation learning on TIGs, researchers have proposed a series of TIG models. However, these models are still facing two tough gaps between the pre-training and downstream predictions in their "pre-train, predict" training paradigm. First, there is a temporal gap that exhibits limited accommodation ability to their timely predictions. This shortcoming severely undermines their applicability in distant future predictions on the dynamically evolving TIG data. Second, there is a semantic gap due to the lack of versatility in these pre-trained models to effectively cater to diverse downstream tasks. This hinders their practical applications, as they struggle to align with their learning and prediction capabilities across various application scenarios. Recently, the "pre-train, prompt" paradigm has emerged as a lightweight mechanism for model generalization. Therefore, applying this paradigm within TIGs is a potential solution to solve the aforementioned challenges. However, the adaptation of this paradigm to TIGs is not straightforward. The prevalent application of prompting in static graph contexts falls short in temporal settings due to a lack of consideration for time-sensitive dynamics and a deficiency in expressive power. To address this issue, we introduce Temporal Interaction Graph Prompting (TIGPrompt), a versatile framework that seamlessly integrates with existing TIG models, bridging both the temporal and semantic gaps mentioned above. In detail, we propose a temporal prompt generator to offer temporally-aware prompts for different tasks. These prompts stand out for their minimalistic design, relying solely on the fine-tuning of the prompt generator with very little supervision data, which is extremely efficient. To cater to varying computational resource demands, we propose an extended "pre-train, prompt-based fine

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Deep Convolutional Neural Network Approach for COVID-19 Detection

引用

computer Systems Science & engineering 2022年第7期42卷 201-211页

作者： Yu Xue Bernard-Marie Onzo Romany F.Mansour Shoubao Su School of Computer and Software Nanjing University of Information Science and TechnologyNanjing210044China Engineering Research Center of Digital Forensics Ministry of EducationNanjing University of Information Science and TechnologyNanjingChina Department of Mathematics Faculty of ScienceNew Valley UniversityEl-Kharja72511Egypt Jiangsu Key Laboratory of Data Science and Smart Software Jinling Institute of TechnologyNanjing211169China

Coronavirus disease 2019(Covid-19)is a life-threatening infectious disease caused by a newly discovered strain of the *** by the end of 2020,Covid-19 is still not fully understood,but like other similar viruses,the main mode of transmission or spread is believed to be through droplets from coughs and sneezes of infected *** accurate detection of Covid-19 cases poses some questions to scientists and *** two main kinds of tests available for Covid-19 are viral tests,which tells you whether you are currently infected and antibody test,which tells if you had been infected ***-tine Covid-19 test can take up to 2 days to complete;in reducing chances of false negative results,serial testing is *** image processing by means of using Chest X-ray images and Computed Tomography(CT)can help radiologists detect the *** imaging approach can detect certain characteristic changes in the lung associated with *** this paper,a deep learning model or tech-nique based on the Convolutional Neural Network is proposed to improve the accuracy and precisely detect Covid-19 from Chest Xray scans by identifying structural abnormalities in scans or X-ray *** entire model proposed is categorized into three stages:dataset,data pre-processing andﬁnal stage being training and classiﬁcation.

关键词： COVID-19 deep learning convolutional neural network X-ray

来源：评论

学校读者我要写书评

暂无评论

Efficient Biomedical Ontology Meta-matching Based on Interpolation Model Based Hybrid Evolutionary Algorithm

Efficient Biomedical Ontology Meta-matching Based on Interpo...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Xue, Xingsi Wu, Qi Ye, Miao Zhu, Hai Huang, Yikun Fujian University of Technology Fujian Provincial Key Laboratory of Big Data Mining and Applications Fuzhou China Taiyuan University of Technology College of Electrical and Power Engineering Taiyuan China Guilin University of Electronic Technology School of Information and Communication Guilin China Henan University of Engineering School of Computer Zhengzhou China Concord University College Fujian Normal University Fuzhou China

ISBN: (纸本)9781665468190

As an advanced biomedical knowledge modeling technology, biomedical ontology models the biomedical domain. However, since the lack of uniform standards for constructing biomedical ontologies, the biomedical ontologies obtained by different construction methods for the same thing may be different, which is known as the biomedical ontology heterogeneity problem. To solve this problem, we need to execute the biomedical ontology matching process, where it is important to integrate different similarity measures to improve the quality of alignment. Evolutionary Algorithm (EA) is an effective algorithm to address the biomedical ontology meta-matching problem. However, the classical EA-based biomedical ontology meta-matching technique needs to traverse reference alignment to evaluate the individuals, which makes the algorithm have high running time. To overcome this drawback, we propose an Interpolation Model (IM) based Hybrid EA (IM-HEA), which combines EA with a problem-specific IM to evaluate the individual and execute the local search process. In particular, we use lattice design to divide the feasible domain into several uniform sub-regions, and on this basis, approximately evaluate the fitness of newly generated individual. In addition, to avoid the algorithm falling into the local optimum, we further introduce an IM-based local search process into EA's evolving process. In the experiment, we test IM-HEA on OAEI's Benchmark and Anatomy and compared them with classical EA in terms of alignment's quality and running time. The experimental results show that IM-HEA greatly enhances the efficiency of EA with little sacrifice on the alignment's quality. © 2022 IEEE.

关键词： Interpolation

来源：评论

学校读者我要写书评

暂无评论

SSC- l0 : Sparse Subspace Clustering with the l0 Inequality Constraint 7th

SSC- l0 : Sparse Subspace Clustering with the l0 Inequality...

引用

7th Asian Conference on Pattern Recognition, ACPR 2023

作者： Wang, Yangbo Zhou, Jie Lin, Qingshui Lu, Jianglin Gao, Can National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518060 China Basic Teaching Department Liaoning Technical University Huludao125105 China College of Engineering Northeastern University BostonMA02115 United States

ISBN: (纸本)9783031476648

Self-expression learning methods often obtain a coefficient matrix to measure the similarity between pairs of samples. However, directly using all points to represent a fixed sample in a class under the self-expression framework may not be ideal, as points from other classes participate in the representing process. To alleviate this issue, this study attempts to achieve representation learning between points only coming from the same class. In practice, it is easier for data points from the same class to represent each other than that from different classes. So, when reconstructing a point, if the number of non-zero elements in the coefficient vector is limited, a model is more likely to select data points from the class where the reconstructed point lies to complete the reconstruction work. Based on this idea, we propose Sparse Subspace Clustering with the l0 inequality constraint (SSC- l0 ). In SSC- l0, the l0 inequality constraint determines the maximum number of non-zero elements in the coefficient vector, which helps SSC- l0 to conduct representation learning among the points in the same class. After introducing the simplex constraint to ensure the translation invariance of the model, an optimization method concerning l0 inequality constraint is formed to solve the proposed SSC- l0, and its convergence is theoretically analyzed. Extensive experiments on well-known datasets demonstrate the superiority of SSC- l0 compared to several state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Neural network based on improved parallel bat algorithm and its application

Journal of Network Intelligence

引用

Journal of Network Intelligence 2021年第3期6卷 428-439页

作者： Zhao, Zhuo-Qiang Liu, Shi-Jian Xu, Lin Pan, Jeng-Shyang Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology Fuzhou350118 China Institute of Artificial Intelligence Fujian University of Technology Fuzhou350118 China STEM University of South Australia Australia College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China

The artificial neural network is a research hotspot emerging in the field of artificial intelligence. Back Propagation (BP) neural network plays an important role in neural network. The network has the disadvantages of slow convergence of learning algorithms and easy falling into a local minimum. In this paper, a parallel bat algorithm with a new communication strategy is proposed, which is used to optimize the weights and thresholds of the BP neural network and establish the improved PBA-BP model. It was applied to optimize the parameters of the PID controller. Finally, we verified the effectiveness of the proposed method through the simulations. © 2021, Taiwan Ubiquitous Information. All rights reserved.

关键词： Three term control systems

来源：评论

学校读者我要写书评

暂无评论

BENCHMARKING COMMUNITY DRUG RESPONSE PREDICTION MODELS: dataSETS, MODELS, TOOLS, AND METRICS FOR CROSS-dataSET GENERALIZATION ANALYSIS

arXiv

引用

arXiv 2025年

作者： Partin, Alexander Vasanthakumari, Priyanka Narykov, Oleksandr Wilke, Andreas Koussa, Natasha Jones, Sara E. Zhu, Yitan Overbeek, Jamie C. Jain, Rajeev Fernando, Gayara Demini Sanchez-Villalobos, Cesar Garcia-Cardona, Cristina Mohd-Yusof, Jamaludin Chia, Nicholas Wozniak, Justin M. Ghosh, Souparno Pal, Ranadip Brettin, Thomas S. Weil, M. Ryan Stevens, Rick L. Division of Data Science and Learning Argonne National Laboratory LemontIL United States Frederick National Laboratory for Cancer Research Cancer Data Science Initiatives Cancer Research Technology Program FrederickMD United States Department of Statistics University of Nebraska–Lincoln LincolnNE United States Department of Electrical & Computer Engineering Texas Tech University LubbockTX United States Division of Computer Computational and Statistical Sciences Los Alamos National Laboratory Los AlamosNM United States Department of Computer Science The University of Chicago ChicagoIL United States

Deep learning (DL) and machine learning (ML) models have shown promise in drug response prediction (DRP), yet their ability to generalize across datasets remains an open question, raising concerns about their real-world applicability. Due to the lack of standardized benchmarking approaches, model evaluations and comparisons often rely on inconsistent datasets and evaluation criteria, making it difficult to assess true predictive capabilities. In this work, we introduce a benchmarking framework for evaluating cross-dataset prediction generalization in DRP models. Our framework incorporates five publicly available drug screening datasets, six standardized DRP models, and a scalable workflow for systematic evaluation. To assess model generalization, we introduce a set of evaluation metrics that quantify both absolute performance (e.g., predictive accuracy across datasets) and relative performance (e.g., performance drop compared to within-dataset results), enabling a more comprehensive assessment of model transferability. Our results reveal substantial performance drops when models are tested on unseen datasets, underscoring the importance of rigorous generalization assessments. While several models demonstrate relatively strong cross-dataset generalization, no single model consistently outperforms across all datasets. Furthermore, we identify CTRPv2 as the most effective source dataset for training, yielding higher generalization scores across target datasets. By sharing this standardized evaluation framework with the community, our study aims to establish a rigorous foundation for model comparison, and accelerate the development of robust DRP models for real-world applications. © 2025, CC BY.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

POI-Enhancer: An LLM-based Semantic Enhancement Framework for POI Representation Learning

arXiv

引用

arXiv 2025年

作者： Cheng, Jiawei Wang, Jingyuan Zhang, Yichuan Ji, Jiahao Zhu, Yuanshao Zhang, Zhibo Zhao, Xiangyu SKLCCSE School of Computer Science and Engineering Beihang University Beijing China Department of Data Science City University of Hong Kong Hong Kong MIIT Key Laboratory of Data Intelligence and Management Beihang University Beijing China School of Economics and Management Beihang University Beijing China

POI representation learning plays a crucial role in handling tasks related to user mobility data. Recent studies have shown that enriching POI representations with multimodal information can significantly enhance their task performance. Previously, the textual information incorporated into POI representations typically involved only POI categories or check-in content, leading to relatively weak textual features in existing methods. In contrast, large language models (LLMs) trained on extensive text data have been found to possess rich textual knowledge. However leveraging such knowledge to enhance POI representation learning presents two key challenges: first, how to extract POI-related knowledge from LLMs effectively, and second, how to integrate the extracted information to enhance POI representations. To address these challenges, we propose POI-Enhancer, a portable framework that leverages LLMs to improve POI representations produced by classic POI learning models. We first design three specialized prompts to extract semantic information from LLMs efficiently. Then, the Dual Feature Alignment module enhances the quality of the extracted information, while the Semantic Feature Fusion module preserves its integrity. The Cross Attention Fusion module then fully adaptively integrates such high-quality information into POI representations and Multi-View Contrastive Learning further injects human-understandable semantic information into these representations. Extensive experiments on three real-world datasets demonstrate the effectiveness of our framework, showing significant improvements across all baseline representations. Copyright © 2025, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Attention-Driven Graph Convolutional Networks for Deadline-Constrained Virtual Machine Task Allocation in Edge Computing

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Ali, Ahmad Ullah, Inam Singh, Sushil Kumar Jiang, Weiwei Alturise, Fahad Bai, Xiaoshan Shenzhen University College of Mechatronics and Control Engineering Shenzhen518060 China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen518060 China Beijing University of Posts and Telecommunications Beijing China Qassim University College of Computer Department of Cybersecurity Buraydah Saudi Arabia

Effective load balancing is critical in contemporary mobile computing environments to ensure optimal resource utilization and adherence to strict deadlines. This study presents a novel framework for deadline-constrained load balancing, termed DCLD-net, which leverages the integration of Graph Convolutional Networks (GCN) and an attention-based Bidirectional Gated Recurrent Unit (BiGRU). The proposed framework utilizes the spatial modeling capabilities of GCNs to capture intricate network topologies, while the temporal dynamics and attention mechanisms of BiGRU enable the identification of critical patterns in traffic flow data. By incorporating key factors such as sticky sessions, load distribution, content-based factors such as videos or file downloads, instance health checks, and other relevant parameters, the DCLD-net model achieves high accuracy in predicting task schedules. These schedules allocate tasks to virtual machines (VMs), enabling dynamic traffic routing through a load balancer that efficiently assigns tasks. Extensive experimental evaluations demonstrate that the proposed attention-based BiGRU-GCN model significantly outperforms traditional methods in both deadline compliance and load balancing efficiency, offering a robust solution to the challenges of modern cloud computing. Experimental results demonstrate that DCLD-net outperforms traditional methods by 27% in deadline compliance and 29.3% in load balancing efficiency. © 1975-2011 IEEE.

关键词： Mobile edge computing

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：