检索结果-内蒙古大学图书馆

2024 International Conference on Cyber-Physical Social Intelligence, ICCSI 2024

作者： Lu, Mengye Zhu, Peng Miao, Chun Wang, Youjian Wang, Jiacun Nanjing University of Science and Technology School of Economics and Management Nanjing China Monmouth University Department of Computer Science and Software Engineering NJ United States

ISBN: (纸本)9798350376739

Group activities are becoming more and more common on the Internet in the big data environment. Which makes many scholars focus on how to recommend items or activities to a group. However, conventional recommendation systems grapple with issues like data sparsity and the cold start problem, which limits the relevance and accuracy of recommendation. Therefore, this paper conducts an in-depth exploration of group recommendation methods utilizing graph neural networks and introduces a novel recommendation framework aimed at enhancing the efficacy of group recommendations. Firstly, this paper discusses how to construct an effective graph structure to encode the social interaction and individual characteristics of users. Then, an improved graph neural network architecture is proposed. Finally, extensive experimental evaluation is carried out on actual datasets, and the findings demonstrate that our framework markedly enhances the accuracy of recommendations. © 2024 IEEE.

关键词： Graph neural network Group recommendation Recommendation system

来源：评论

学校读者我要写书评

暂无评论

Super-resolution reconstruction of single image for latent features

引用

Computational Visual Media 2024年第6期10卷 1219-1239页

作者： Xin Wang Jing-Ke Yan Jing-Ye Cai Jian-Hua Deng Qin Qin Yao Cheng School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610000 China State Key Laboratory of Rail Transit Vehicle System Southwest Jiaotong UniversityChengdu 610000China School of Marine Engineering Guilin University of Electronic TechnologyGuilin 541000China School of Computer and Information Security Guilin University of Electronic TechnologyGuilin 541000China

Single-image super-resolution(SISR)typically focuses on restoring various degraded low-resolution(LR)images to a single high-resolution(HR)***,during SISR tasks,it is often challenging for models to simultaneously maintain high quality and rapid sampling while preserving diversity in details and texture *** challenge can lead to issues such as model collapse,lack of rich details and texture features in the reconstructed HR images,and excessive time consumption for model *** address these problems,this paper proposes a Latent Feature-oriented Diffusion Probability Model(LDDPM).First,we designed a conditional encoder capable of effectively encoding LR images,reducing the solution space for model image reconstruction and thereby improving the quality of the reconstructed *** then employed a normalized flow and multimodal adversarial training,learning from complex multimodal distributions,to model the denoising *** so boosts the generative modeling capabilities within a minimal number of sampling *** comparisons of our proposed model with existing SISR methods on mainstream datasets demonstrate that our model reconstructs more realistic HR images and achieves better performance on multiple evaluation metrics,providing a fresh perspective for tackling SISR tasks.

关键词： image superresolution reconstruction denoising diffusion probabilistic model normalized flow adversarial neural network variational auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Boosting the Prediction of Brain Tumor Using Two Stage BiGait Architecture 25

Boosting the Prediction of Brain Tumor Using Two Stage BiGai...

引用

25th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2024

作者： Khan, Saif-Ur-Rehman Khan, Zia Hossain, Md Zakir Mayumu, Nicanor Yasmin, Farhana Aziz, Younas School of Computer Science and Engineering Central South University Changsha China Curtin University Perth Australia Nanjing University of information Science and Technology Department of Computer and Software Nanjing China

ISBN: (纸本)9798350379037

Deep learning (DL) techniques hold immense promise for revolutionizing medical diagnostics, including brain tumor detection. Detecting malignancies in the brain is fraught with challenges that carry critical implications for accurate and timely diagnosis. Early detection is paramount, and our paper presents a novel approach to address this issue. We introduce the BiGait ensemble transfer learning (TL) architecture tailored for brain tumor classification tasks. Our proposed methodology is a two-stage process that harnesses the power of TL. In the initial phase, we employ two established models, ResNet50 and EfficientNetB1, and fine-tune them using the Kaggle multiclass dataset. This includes the incorporation of supplementary feature layers. We also leverage a stack ensemble solution to combine these feature layers, enhancing the model's overall predictive capabilities. Our approach is designed to shift the input distribution positively, effectively mitigating overfitting and reducing hardware overhead. To validate the effectiveness of our method, we conducted extensive experiments and evaluation. The dataset utilized in this work, publicly available on Kaggle contains 3,264 brain MRIs categorized into four groups: normal (500 images), meningioma (937 images), glioma (926 images), and pituitary tumor (901 images). The experimental results demonstrate that our model surpasses traditional single-model approaches, achieving an impressive accuracy rate of 96% and exhibiting superior generalization capabilities. This advancement in brain tumor classification holds great promise for early diagnosis and improved patient outcomes. © 2024 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Minimizing the cost of periodically replicated systems via model and quantitative analysis

引用

Frontiers of computer science 2024年第5期18卷 11-24页

作者： Chenhao ZHANG Liang WANG Limin XIAO Shixuan JIANG Meng HAN Jinquan WANG Bing WEI Guangjun QIN State Key Laboratory of Software Development Environment Beihang UniversityBeijing 100191China School of Computer Science and Engineering Beihang UniversityBeijing 100191China School of Cyberspace Security Hainan UniversityHaikou 570228China Smart City College Beijing Union UniversityBeijing 100101China

Geographically replicating objects across multiple data centers improves the performance and reliability of cloud storage *** consistent replicas comes with high synchronization costs,as it faces more expensive WAN transport prices and increased *** replication is the widely used technique to reduce the synchronization *** replication strategies in existing cloud storage systems are too static to handle traffic changes,which indicates that they are inflexible in the face of unforeseen loads,resulting in additional synchronization *** propose quantitative analysis models to quantify consistency and synchronization cost for periodically replicated systems,and derive the optimal synchronization period to achieve the best tradeoff between consistency and synchronization *** on this,we propose a dynamic periodic synchronization method,Sync-Opt,which allows systems to set the optimal synchronization period according to the variable load in clouds to minimize the synchronization *** results demonstrate the effectiveness of our *** with the policies widely used in modern cloud storage systems,the Sync-Opt strategy significantly reduces the synchronization cost.

关键词： periodic replication consistency maintenance synchronization cost synchronization strategy

来源：评论

学校读者我要写书评

暂无评论

An end-to-end contrastive deep-learning framework for remote physiological signal measurement 3

An end-to-end contrastive deep-learning framework for remote...

引用

3rd Vision-Based Remote Physiological Signal Sensing Challenge and Workshop, RePSS 2024

作者： Wu, Bingjie Zhou, Menghan Liu, Wei Wang, Xingyao Zheng, Xingjian Xie, Yiping Luo, Chaoqi Zhen, Liangli Singapore Singapore College of Computer Science and Software Engineering Shenzhen University Shenzhen China School of Electrical Engineering Southwest Jiaotong University Chengdu China

Heart rate measurements based on remote physiological signals could significantly facilitate health monitoring in daily life. However, the ground-truth labels of the physiological signals are expensive and hard to collect. In this paper, we present a contrastive self-supervised learning framework to extract discriminative remote physiological features by leveraging periodic signal priors without ground-truth labels in the pre-training stage. Specifically, a ranking loss and a contrastive learning loss are constructed to extract knowledge with resampling of the video clips. In addition, data augmentation and ensemble learning strategies are designed to fine-tune the pre-trained model and fuse the results to improve the heart rate measurement. Our final solution achieves the 1st place in track 1 of the 3rd Vision-based Remote Physiological Signal Sensing (RePSS) Challenge. © 2024 Copyright for this paper by its authors.

关键词： contactless heart rate measurement Remote photoplethysmography RePSS self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Imitation Learning from Purified Demonstrations 41

Imitation Learning from Purified Demonstrations

引用

41st International Conference on Machine Learning, ICML 2024

作者： Wang, Yunke Dong, Minjing Zhao, Yukun Du, Bo Xu, Chang School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Wuhan Institute of Data Intelligence Wuhan University China Department of Computer Science City University of Hong Kong Hong Kong School of Computer Science Faculty of Engineering The University of Sydney Australia

Imitation learning has emerged as a promising approach for addressing sequential decision-making problems, with the assumption that expert demonstrations are optimal. However, in real-world scenarios, most demonstrations are often imperfect, leading to challenges in the effectiveness of imitation learning. While existing research has focused on optimizing with imperfect demonstrations, the training typically requires a certain proportion of optimal demonstrations to guarantee performance. To tackle these problems, we propose to purify the potential noises in imperfect demonstrations first, and subsequently conduct imitation learning from these purified demonstrations. Motivated by the success of diffusion model, we introduce a two-step purification via diffusion process. In the first step, we apply a forward diffusion process to smooth potential noises in imperfect demonstrations by introducing additional noise. Subsequently, a reverse generative process is utilized to recover the optimal demonstration from the diffused ones. We provide theoretical evidence supporting our approach, demonstrating that the distance between the purified and optimal demonstration can be bounded. Empirical results on MuJoCo and RoboSuite demonstrate the effectiveness of our method from different aspects. 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

A Fine-Grained Detector of Face Mask Wearing Status Based on Improved YOLOX

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第4期5卷 1816-1830页

作者： Xiao, Hongli Wang, Bingshu Zheng, Jiangbin Liu, Licheng Philip Chen, C.L. Northwestern Polytechnical University School of Software Taicang Campus Suzhou215400 China Shanghai Jiaotong University School of Electronic Information and Electrical Engineering Shanghai200240 China Hunan University College of Electrical and Information Engineering Changsha410082 China South China University of Technology School of Computer Science and Engineering Guangzhou510641 China

The fast outbreak of coronavirus disease 2019 (COVID-19) and rapid proliferation of its variants have continued to pose a huge challenge to people around the world. Wearing medical masks properly in public and private settings can protect people from COVID-19, which brings a growing demand for automatic detection services of face mask wearing conditions. In this article, we propose a fine-grained detector called ECA_YOLOX-S to identify the wearing status of face masks. Efficient channel attention is introduced into YOLOX-S to reach a tradeoff between effectiveness and efficiency. To demonstrate the performance of our proposed method, a Fine-grained Face Mask (FineFM) dataset is created, which covers four classes of mask wearing status. The proposed FineFM dataset has 16 955 annotated images and covers multiple realistic scenarios. To the best of authors' knowledge, it has the largest number of improper mask wearing images among all similar datasets for realistic scenes. Experiments conducted on the FineFM dataset demonstrate that ECA_YOLOX-S achieves an overall mean average precision (mAP)@.50:95 of 86.80% for moderate scenes and an overall mAP@.50:95 of 73.20% for complex scenes, outperforming its benchmark model. Moreover, experiments conducted on other realistic and simulated datasets indicate that the proposed detector has advantages over other methods. © 2020 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

UAV Communication and Localization Network Deployment for Content Distribution in Emergency Scenarios 6

UAV Communication and Localization Network Deployment for Co...

引用

6th International Conference on Electronics and Communication, Network and computer Technology, ECNCT 2024

作者： Wang, Kailiang Yu, Ran Zhang, Donghui Zhang, Yuhao Lu, Jie Yu, Xinyu Men, Baoxia Zhang, Yi Ji, Yutong State Grid Jibei Information & Telecommunication Company Beijing China School of Computer Science National Pilot Software Engineering School Bupt Beijing China

ISBN: (纸本)9798350366143

Unmanned Aerial Vehicles (UAVs) are increasingly recognized for their potential to revolutionize emergency response communications and localization, especially when traditional infrastructure is damaged or non-existent. This study presents an iterative deployment strategy for UAVs equipped with directional antennas to optimize communication and localization in disaster-stricken areas. By integrating a line-of-sight coverage model with real-world environmental data, and utilizing a greedy algorithm alongside an improved Particle Swarm Optimization algorithm, the strategy refines UAV placement and antenna orientations iteratively. The results of simulations show that our method significantly enhances communication rates and localization accuracy, proving effective in complex emergency scenarios. © 2024 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Self-adaptive label filtering learning for unsupervised domain adaptation

引用

Frontiers of computer science 2023年第1期17卷 225-227页

作者： Qing TIAN Heyang SUN Shun PENG Tinghuai MA School of Computer and Software Nanjing University of Information Science and TechnologyNanjing 210044China Engineering Research Center of Digital Forensics Ministry of EducationNanjing University of Information Science and TechnologyNanjing 210044China

1 Introduction As an emerging machine learning paradigm,unsupervised domain adaptation(UDA)aims to train an effective model for unlabeled target domain by leveraging knowledge from related but distribution-inconsistent source *** of the existing UDA methods[2]align class-wise distributions resorting to target domain pseudo-labels,for which hard labels may be misguided by misclassifications while soft labels are confusing with trivial noises so that both of them tend to cause frustrating *** overcome such drawbacks,as shown in Fig.1,we propose to achieve UDA by performing self-adaptive label filtering learning(SALFL)from both the statistical and the geometrical perspectives,which filters out the misclassified pseudo-labels to reduce negative ***,the proposed SALFL firstly predicts labels for the target domain instances by graph-based random walking and then filters out those noise labels by self-adaptive learning strategy.

关键词： filtering resort overcome

来源：评论

学校读者我要写书评

暂无评论

Real-Time DAG Task Allocation Strategy for Multiprocessor by Optimistic Parallelism 24

Real-Time DAG Task Allocation Strategy for Multiprocessor by...

引用

24th IEEE International Conference on Communication Technology, ICCT 2024

作者： Chen, Yumeng Liu, Songlin Zhou, Jianhong Ling, Xiang University of Electronic Science and Technology of China National Key Laboratory of Science and Technology on Communications China Xihua University School of Computer and Software Engineering China

ISBN: (纸本)9798350363760

In modern real-time operating systems, complex task loads are often modeled as directed acyclic graphs (DAG) and executed in parallel on multiprocessor systems. The topological constraints present in DAG tasks prevent the DAG task allocation strategy from fully leveraging the parallel advantage of multiprocessor platforms. We analyze the parallelism of DAG tasks and propose a method to limit the number of processors that can be executed by DAG tasks. This method provides more idle processors and processor time for high parallelism DAG tasks by limiting the number of processors that can be executed by low parallelism DAG tasks. In addition, we develope optimistic parallelism scheduling algorithm (OPSA) based on the remaining utilization with low complexity. The experimental results show that our DAG task allocation strategy fully utilizes processor resources, reduces interference between DAG tasks, and achieves better worst case response time (WCRT) analysis results and shorter actual task's WCRT on real-time multiprocessor system. © 2024 IEEE.

关键词： Directed graphs

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：