检索结果-内蒙古大学图书馆

International Conference on Artificial Intelligence in Information and Communication (ICAIIC)

作者： Yunseong Lee Donghyun Lee Junsuk Oh Chihyun Song Wonjong Noh Sungrae Cho School of Computer Science and Engineering Chung-Ang University Seoul Republic of Korea School of Software Hallym University Chuncheon Republic of Korea

ISBN: (数字)9798331507022

ISBN: (纸本)9798331507039

This paper explores reinforcement learning (RL) based on resource allocation in cell-free networks, a promising alternative to traditional cellular architectures. Cell-free networks eliminate cell boundaries by using distributed access points (APs) to collaboratively serve users, improving spectral efficiency and service uniformity. However, managing power allocation, beamforming, and AP clustering in such decentralized environments presents new challenges. We present several RL-based algorithms aimed at optimizing key network functions. Specifically, the study explores dynamic power control, advanced beamforming techniques, and multi-agent RL frameworks for clustering access points. The proposed methods leverage the adaptability of RL to optimize network performance in real-time under varying conditions.

关键词： Wireless communication Array signal processing Spectral efficiency Microprocessors Refining Power control Reinforcement learning computer architecture Real-time systems Resource management

来源：评论

学校读者我要写书评

暂无评论

Decentralized Federated Graph Learning with Lightweight Zero Trust Architecture for Next-Generation Networking Security

引用

IEEE Journal on Selected Areas in Communications 2025年第6期43卷 1908-1922页

作者： Zhou, Xiaokang Liang, Wei Wang, Kevin I-Kai Yada, Katsutoshi Yang, Laurence T. Ma, Jianhua Jin, Qun Kansai University Faculty of Business Data Science Osaka565-0823 Japan RIKEN Center for Advanced Intelligence Project RIKEN Tokyo103-0027 Japan Hunan University of Technology and Business School of Computer Science Changsha410205 China Xiangjiang Laboratory Changsha410205 China The University of Auckland Department of Electrical Computer and Software Engineering Auckland1010 New Zealand St. Francis Xavier University Department of Computer Science AntigonishNSB2G 2W5 Canada Hosei University Faculty of Computer and Information Sciences Chiyoda-ku102-8160 Japan Waseda University Faculty of Human Sciences Tokorozawa359-1192 Japan

The rapid development and usage of digital technologies in modern intelligent systems and applications bring critical challenges on data security and privacy. It is essential to allow cross-organizational data sharing to achieve smart service provisioning, while preventing unauthorized access and data leak to ensure end users' efficient and secure collaborations. Federated Learning (FL) offers a promising pathway to enable innovative collaboration across multiple organizations. However, more stringent security policies are needed to ensure authenticity of participating entities, safeguard data during communication, and prevent malicious activities. In this paper, we propose a Decentralized Federated Graph Learning (FGL) with Lightweight Zero Trust Architecture (ZTA) model, named DFGL-LZTA, to provide context-aware security with dynamic defense policy update, while maintaining computational and communication efficiency in resource-constrained environments, for highly distributed and heterogeneous systems in next-generation networking. Specifically, with a re-designed lightweight ZTA, which leverages adaptive privacy preservation and reputation-based aggregation together to tackle multi-level security threats (e.g., data-level, model-level, and identity-level attacks), a Proximal Policy Optimization (PPO) based Deep Reinforcement Learning (DRL) agent is introduced to enable the real-time and adaptive security policy update and optimization based on contextual features. A hierarchical Graph Attention Network (GAT) mechanism is then improved and applied to facilitate the dynamic subgraph learning in local training with a layer-wise architecture, while a so-called sparse global aggregation scheme is developed to balance the communication efficiency and model robustness in a P2P manner. Experiments and evaluations conducted based on two open-source datasets and one synthetic dataset demonstrate the usefulness of our proposed model in terms of training performance, computa

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Spatiotemporal Feature Extraction and Fusion for Longitudinal Alzheimer’s Disease Diagnosis 17th

Spatiotemporal Feature Extraction and Fusion for Longitudina...

引用

17th International Conference on Brain Informatics, BI 2024

作者： Guan, Zhenghua Yang, Peng Lei, Haijun Yang, Bao Song, Xuegang Dong, Lei Yan, Xueqin Wei, Cuimei Liang, Chunhua Xiao, Xiaohua Wang, Tianfu Lei, Baiying School of Biomedical Engineering Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging National-Regional Key Technology Engineering Laboratory for Medical Ultrasound Shenzhen University Medical schoo1 Shenzhen518060 China Affiliated Hospital of Shenzhen University Health Science Center Shenzhen University Shenzhen Second People’s Hospital Shenzhen China School of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China

ISBN: (纸本)9789819632961

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by a gradual decline in cognitive function. The brain structure of AD patients progressively atrophies over time. Longitudinal data can capture the changes in imaging over time, providing important information for early detection and diagnosis of the disease, which is not achievable with images from a single time point. We propose a novel longitudinal AD diagnosis model that leverages deformation field data obtained by registering images from adjacent time points. Each pixel in the deformation field represents the displacement vector of the corresponding pixel between two scans, reflecting information about brain tissue atrophy. By effectively extracting both temporal and spatial features from the longitudinal data, our model enhances early AD diagnosis. The input data for the model consists of processed sMRI images and deformation fields calculated between each pair of time points in the longitudinal sMRI data. The model is composed of three modules: Convolutional Long Short-Term Memory (CLSTM), Gate Fusion Long Short-Term Memory (GFLSTM), and an attention module. CLSTM extracts temporal features while preserving spatial features by maintaining the 3D shape of the features, GFLSTM achieves spatiotemporal feature fusion within a cascaded network, and the attention module further enhances feature representation. Our proposed method was evaluated on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, and experiments demonstrate that it outperforms several state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

High-Fidelity Diffusion Editor for Zero-Shot Text-Guided Video Editing

High-Fidelity Diffusion Editor for Zero-Shot Text-Guided Vid...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Yan Luo Zhichao Zuo Zhao Zhang Zhongqiu Zhao Haijun Zhang Richang Hong School of Computer Science and Information Engineering Hefei University of Technology Hefei China Yunnan Key Laboratory of Software Engineering Yunan China School of Computer Science Harbin Institute of Technology Shenzhen China

ISBN: (数字)9798331506681

ISBN: (纸本)9798331506698

Text-guided image generative diffusion models achieve fast development on the generation and editing of high-quality images. To extend such success to video editing, some efforts combining image generation with video editing have been made, which however only achieve inferior performance. We attribute it to two challenges: 1) different from the static image generation, it is tricky for dynamic video information to ensure the temporal fidelity of motion consistency across different frames; 2) the randomness of the frame generation process makes it hard to continuously retain the similar spatial fidelity for the original detailed features. In this paper, we propose a new high-fidelity diffusion model-based zero-shot text-guided video editing network, called HiFiVEditor, which aims to conduct effective video editing with high fidelity of the original video's detailed and dynamic information. Specifically, we propose a Spatial-Temporal Fidelity Block (STFB) that enables the model to restore the spatial features by enlarging the spatial perceptual field to avoid loss of important information, and capture more dynamic information between different frames by using all frames for preserving temporal consistency to achieve better temporal fidelity. In addition, we introduce Null-Text Embedding to create a soft text embedding to optimize the noise learning process, so that the latent noise can be aligned with the prompt. Furthermore, to tune the video style and render it more realistic, we employ a Prior-Guided Perceptual Loss to constrain the prediction results to avoid deviating from the original video style. Extensive experiments demonstrate the superior video editing capability compared to existing works.

关键词： Image synthesis Noise Dynamics Diffusion models Data models Image restoration Data mining

来源：评论

学校读者我要写书评

暂无评论

Blended Learning Application in the Principle of computer Composition 2021

Blended Learning Application in the Principle of Computer Co...

引用

5th International Conference on E-Society, E-Education and E-Technology, ICSET 2021

作者： Peng, Shih Jung School of Computer Science and Software Engineering Zhaoqing University 86-*** China

ISBN: (纸本)9781450390156

Blended learning is always a significant method of learning and teaching in education. This method combines the advantages of online and face-to-face learning models to improve students' interest in learning and teaching results. Because the course ĝ€Principle of computer Composition' is extensive, which is not easy to understand and absorb in the process of learning, students need more basic knowledge in learning. Therefore, we establish a blended learning platform for the course, the factors such as learning courses, learning media, and learning environment would be optimized. Utilizing blended learning modules, for instance, ĝ€e-learning platform', ĝ€online class group', ĝ€face-to-face classes', ĝ€validation in experiments', ĝ€discussion of course content', and ĝ€learning examination', we can optimize the teaching process and effect, and improve students' interest in learning and learning impact. At the end of the blended learning module, we feedback the student's learning data to the teacher who obtains the distribution of students' academic achievement through statistical analysis as a reference for adjusting teaching content, progress, and methods. © 2021 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

CAPAST: Content Affinity Preserved Arbitrary Style Transfer

CAPAST: Content Affinity Preserved Arbitrary Style Transfer

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xinyuan Zheng Xiaojie Li Canghong Shi Jia He Zhan Ao Huang Xian Zhang Imran Mumtaz School of Computer Science Chengdu University of Information Technology China School of Computer and Software Engineering Xihua University China Faculty of Engineering and Mathematical Sciences The University of Western Australia Australia Department of Computer Science University of Agriculture Faisalabad Pakistan

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Balancing the consistency of style and the integrity of content is the main challenge in arbitrary style transfer domain. Currently, local style details can be effectively captured by attention mechanism but easily produce distorted style patterns and inconsistent content structure. In this paper, we propose a Content Affinity Preserving Arbitrary Style Transfer (CAPAST) framework to ensure style features can be stably integrated into the content structure. Considering the local feature learning ability of CNN and the global feature representation advantage of transformer, a dual encoder is proposed to capture local and global features of images with the combination between transformer and CNN. In addition, a channel and spatially aligned attention (CSAA) is introduced to generate high-quality results by stably fusing style features and content features. In experiments, we demonstrated the superior performance of our method in preventing content structure distortion and maintaining consistency between style and content. Codes are available at https://***/miaopashi-zxy/CAPAST.

关键词： Representation learning Acoustic distortion Codes Attention mechanisms Fuses Transformers Acoustics Speech processing

来源：评论

学校读者我要写书评

暂无评论

Researches on Resource Allocation With Reinforcement Learning in Cell-Free Networks

Researches on Resource Allocation With Reinforcement Learnin...

引用

7th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2025

作者： Lee, Yunseong Lee, Donghyun Oh, Junsuk Song, Chihyun Noh, Wonjong Cho, Sungrae School of Computer Science and Engineering Chung-Ang University Seoul Korea Republic of School of Software Hallym University Chuncheon Korea Republic of

ISBN: (纸本)9798331507022

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

CBLA: Empowering Virtual Sensor Nodes with Zero Deployment Costs for SHM Systems

CBLA: Empowering Virtual Sensor Nodes with Zero Deployment C...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Yang Wang Ke Lei Chong Zhang Xin Wang Xin Shi Aihua Deng School of Computer Science and Software Engineering Southwest Petroleum University Chengdu China School of Mechanical Engineering Southwest Petroleum University Chengdu China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Structural Health Monitoring (SHM) monitors the working states of building structures to ensure their reliable operation, but the high cost of SHM sensor nodes limits its largescale deployment. In this paper, we propose a novel computational model that can generate "virtual" sensor nodes with reliable data output at zero deployment costs for cost-efficient SHM sensing. To achieve this, we combine one-dimensional convolutional layer in CNN with Bi-LSTM model to capture temporal and spatial correlations in the data; to improve the quality of the generated data, we design a weighted smoothing algorithm to reduce noise while preserving the integrity of the data; to enhance system robustness, we integrate attention mechanisms at the end of the model to assign different weights to each element, which can capture the limited but crucial information within the data. We evaluate our system on the dataset of the KW51 railroad bridge as an example. The results show that the generated virtual sensor node incurs zero deployment cost, and its data is 94% close to the ground truth, thus making it helpful for facilitating the largescale deployment of SHM systems.

关键词： Costs Smoothing methods Correlation Soft sensors Noise Neural networks Data models

来源：评论

学校读者我要写书评

暂无评论

NUWADYNAMICS: DISCOVERING AND UPDATING IN CAUSAL SPATIO-TEMPORAL MODELING 12

NUWADYNAMICS: DISCOVERING AND UPDATING IN CAUSAL SPATIO-TEMP...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Wang, Kun Wu, Hao Duan, Yifan Zhang, Guibin Wang, Kai Peng, Xiaojiang Zheng, Yu Liang, Yuxuan Wang, Yang China Suzhou Institute for Advanced Research USTC China School of Software Engineering USTC China School of Computer Science USTC China Hong Kong Tongji University China Shenzhen Technology University China National University of Singapore Singapore JD iCity JD Technology

Spatio-temporal (ST) prediction plays a pivotal role in earth sciences, such as meteorological prediction, urban computing. Adequate high-quality data, coupled with deep models capable of inference, are both indispensable and prerequisite for achieving meaningful results. However, the sparsity of data and the high costs associated with deploying sensors lead to significant data imbalances. Models that are overly tailored and lack causal relationships further compromise the generalizabilities of inference methods. Towards this end, we first establish a causal concept for ST predictions, named NuwaDynamics, which targets to identify causal regions in data and endow model with causal reasoning ability in a two-stage process. Concretely, we initially leverage upstream self-supervision to discern causal important patches, imbuing the model with generalized information and conducting informed interventions on complementary trivial patches to extrapolate potential test distributions. This phase is referred to as the discovery step. Advancing beyond the discovery step, we transfer the data to downstream tasks for targeted ST objectives, aiding the model in recognizing a broader potential distribution and fostering its causal perceptual capabilities (denoted as Update step). Our concept aligns seamlessly with the contemporary backdoor adjustment mechanism in causality theory. Extensive experiments on six real-world ST benchmarks showcase that models can gain outcomes upon the integration of the NuwaDynamics concept. NuwaDynamics also can significantly benefit a wide range of changeable ST tasks like extreme weather and long temporal step super-resolution predictions. Our codes are available at https://***/easylearningscores/NuwaDynamics. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Computation theory

来源：评论

学校读者我要写书评

暂无评论

AI-Enhanced threat analysis and risk assessment for connected and autonomous vehicles

引用

software Quality Journal 2025年第3期33卷 1-19页

作者： Ahmad, Usman Han, Mu Mahmood, Shahid School of Computer Science and Communication Engineering Jiangsu University Zhenjiang China The University of Faisalabad Faisalabad Pakistan Intelligent Mobility and Software FEV Iberia Barcelona Spain

Threat Analysis and Risk Assessment (TARA) methodologies aid in the early detection of potential threats during the concept and design phases of automotive security engineering process. Their importance is underscored by the relevant industry standards such ISO/SAE 21434 and SAE J3061. While human expertise and involvement are vital for conducting effective TARA, Artificial Intelligence (AI) can enhance the process significantly by automating certain aspects and thus streamlining operations. This study introduces an AI-driven approach to enhance the implementation of TARA. The proposed method leverages machine learning and evolutionary computing algorithms to identify cybersecurity threats in Connected and Autonomous Vehicles (CAVs) through the STRIDE framework. A structured and preprocessed dataset, containing various threat scenarios and characteristics specific to CAVs, facilitates the application of these AI techniques. The automation focuses on the threat identification and assessment phases of TARA. Machine learning algorithms KNN and Naïve Bayes are evaluated using metrics like accuracy, precision, recall, F1 score, and confusion matrix, achieving 87.5% accuracy. The Ant Colony Optimization (ACO) algorithm predicts attack feasibility ratings, confirming the model’s effectiveness. These metrics demonstrate the AI models’ ability to identify and analyze cybersecurity threats. The results show AI’s potential to transform automotive cybersecurity by automating key TARA stages and providing an efficient framework for threat analysis and risk evaluation.

关键词：

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：