检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Gain, Mrityunjoy Kim, Kitae Raha, Avi Deb Adhikary, Apurba Saad, Walid Han, Zhu Hong, Choong Seon Department of Artificial Intelligence School of Computing Kyung Hee University Yongin17104 Korea Republic of Department of Computer Science and Engineering School of Computing Kyung Hee University Yongin17104 Korea Republic of Department of Computer Science and Engineering School of Computing Kyung Hee University Yongin-si17104 Korea Republic of Department of Information and Communication Engineering Noakhali Science and Technology University Noakhali3814 Bangladesh Bradley Department of Electrical and Computer Engineering Virginia Tech ArlingtonVA22203 United States Electrical and Computer Engineering Department University of Houston HoustonTX77004 United States

Next-generation (NextG) networks will adopt cloud-based, disaggregated designs with open interfaces, thereby enabling tailored data-driven control. The open radio access network (ORAN) paradigm enhances RAN optimization by enabling intelligent and flexible network management. The ORAN Alliance standardizes open and virtualized architectures, supporting services such as enhanced mobile broadband (eMBB), ultra-reliable low-latency communication (uRLLC), and massive machine-type communication (mMTC). Traditionally, eMBB, uRLLC, and mMTC services are tied to specific use equipment (UE) optimized for one service. NextG networks, however, aim to enable multiple services on a single device, crucial for applications like the metaverse. To fill this, in this paper, an artificial intelligence (AI)driven efficient RAN management framework is proposed. This framework introduces the concept of the multi-service-modal UE (MSMU) system, which allows a single UE to handle both eMBB and uRLLC services. The proposed framework integrates traffic demand prediction, route optimization, RAN slicing, service identification, and radio resource management under uncertainty. The challenge of dynamic environments in such a system is addressed by decomposing the optimization problem into long-term (L-SP) and short-term (S-SP) subproblems. Using a long short-term memory (LSTM) model, the proposed approach allows the prediction of eMBB and uRLLC traffic demands and optimal routes for RAN slicing in the L-SP. For the S-SP, another LSTM model is employed to handle real-time service type identification and resource management based on long-term predictions. To support continuous adaptation, continual learning is incorporated into the S-SP framework, allowing the model to learn new service types while retaining prior knowledge. Experimental results show that the proposed framework efficiently manages dual-mode UEs, achieving low mean square error for traffic demand (0.003), resource block prediction

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

MAPS: A Noise-Robust Progressive Learning Approach for Source-Free Domain Adaptive Keypoint Detection

arXiv

引用

arXiv 2023年

作者： Ding, Yuhe Liang, Jian Jiang, Bo Zheng, Aihua He, Ran The School of Computer Science and Technology Anhui University China The School of Artificial Intelligence Anhui University China The Center for Research on Intelligent Perception and Computing CASIA China

Existing cross-domain keypoint detection methods always require accessing the source data during adaptation, which may violate the data privacy law and pose serious security concerns. Instead, this paper considers a realistic problem setting called source-free domain adaptive keypoint detection, where only the well-trained source model is provided to the target domain. For the challenging problem, we first construct a teacher-student learning baseline by stabilizing the predictions under data augmentation and network ensembles. Built on this, we further propose a unified approach, Mixup Augmentation and Progressive Selection (MAPS), to fully exploit the noisy pseudo labels of unlabeled target data during training. On the one hand, MAPS regularizes the model to favor simple linear behavior in-between the target samples via self-mixup augmentation, preventing the model from over-fitting to noisy predictions. On the other hand, MAPS employs the self-paced learning paradigm and progressively selects pseudo-labeled samples from ‘easy’ to ‘hard’ into the training process to reduce noise accumulation. Results on four keypoint detection datasets show that MAPS outperforms the baseline and achieves comparable or even better results in comparison to previous non-source-free counterparts. The code will be released at https://***/YuheD/MAPS. Copyright © 2023, The Authors. All rights reserved.

关键词： Data privacy

来源：评论

学校读者我要写书评

暂无评论

Gradformer: Graph Transformer with Exponential Decay

arXiv

引用

arXiv 2024年

作者： Liu, Chuang Yao, Zelin Zhan, Yibing Ma, Xueqi Pan, Shirui Hu, Wenbin School of Computer Science Wuhan University Wuhan China JD Explore Academy *** China School of Computing and Information Systems The University of Melbourne Melbourne Australia School of Information and Communication Technology Griffith University Brisbane Australia

Graph Transformers (GTs) have demonstrated their advantages across a wide range of tasks. However, the self-attention mechanism in GTs overlooks the graph's inductive biases, particularly biases related to structure, which are crucial for the graph tasks. Although some methods utilize positional encoding and attention bias to model inductive biases, their effectiveness is still suboptimal analytically. Therefore, this paper presents Gradformer, a method innovatively integrating GT with the intrinsic inductive bias by applying an exponential decay mask to the attention matrix. Specifically, the values in the decay mask matrix diminish exponentially, correlating with the decreasing node proximities within the graph structure. This design enables Gradformer to retain its ability to capture information from distant nodes while focusing on the graph's local details. Furthermore, Gradformer introduces a learnable constraint into the decay mask, allowing different attention heads to learn distinct decay masks. Such an design diversifies the attention heads, enabling a more effective assimilation of diverse structural information within the graph. Extensive experiments on various benchmarks demonstrate that Gradformer consistently outperforms the Graph Neural Network and GT baseline models in various graph classification and regression tasks. Additionally, Gradformer has proven to be an effective method for training deep GT models, maintaining or even enhancing accuracy compared to shallow models as the network deepens, in contrast to the significant accuracy drop observed in other GT *** are available at https://***/LiuChuang0059/Gradformer. Copyright © 2024, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

A one-dimensional model for axisymmetric deformations of an inflated hyperelastic tube of finite wall thickness

arXiv

引用

arXiv 2023年

作者： Yu, Xiang Fu, Yibin School of Computer Science and Technology Dongguan University of Technology Dongguan China School of Computing and Mathematics Keele University StaffordshireST5 5BG United Kingdom

We derive a one-dimensional (1d) model for the analysis of bulging or necking in an inflated hyperelastic tube of finite wall thickness from the three-dimensional (3d) finite elasticity theory by applying the dimension reduction methodology proposed by Audoly and Hutchinson (J. Mech. Phys. Solids, 97, 2016). The 1d model makes it much easier to characterize fully nonlinear axisymmetric deformations of a thick-walled tube using simple numerical schemes such as the finite difference method. The new model recovers the diffuse interface model for analyzing bulging in a membrane tube and the 1d model for investigating necking in a stretched solid cylinder as two limiting cases. It is consistent with, but significantly refines, the exact linear and weakly nonlinear bifurcation analyses. Comparisons with finite element simulations show that for the bulging problem, the 1d model is capable of describing the entire bulging process accurately, from initiation, growth, to propagation. The 1d model provides a stepping stone from which similar 1d models can be derived and used to study other effects such as anisotropy and electric loading, and other phenomena such as rupture. Copyright © 2023, The Authors. All rights reserved.

关键词： Finite difference method

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Edge Service Deployment for Cyber Physical Social Intelligence in Internet of Vehicles

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2023年 1-13页

作者： Meng, Ke Liu, Zhongjian Xu, Xiaolong Xia, Xiaoyu Tian, Hao Qi, Lianyong Zhou, Xiaokang School of Computer Science Nanjing University of Information Science and Technology Nanjing China School of Software Nanjing University of Information Science and Technology Nanjing China School of Computing Technologies RMIT University Melbourne Australia State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Qingdao China Faculty of Data Science Shiga University Otsu Japan

Cyber Physical Social Intelligence (CPSI) integrates the social intelligence and cyber-physical systems, enabling machines to better interact and respond to human social behaviors. Under CPSI, the Internet of Vehicles (IoV) has given rise to an increasing number of latency-sensitive services. Edge computing, as a distributed computing paradigm, enhances data processing capabilities, reduces data transmission latency, and minimizes bandwidth consumption. However, due to the limited resources of edge servers, striking a balance between service latency and deployment costs remains a highly challenging issue in the process of service deployment. In this paper, we propose a heterogeneous edge service deployment method for CPSI in IoV. Firstly, considering the heterogeneity of IoV services and edge servers, communication model, computational model, and heterogeneous service deployment cost model are constructed. Secondly, to maximize service deployment efficiency and minimize communication latency, a distance and workload-based edge server cluster division method is proposed. Subsequently, heterogeneous service deployment is performed in different clusters based on service category prioritization and minimal deployment quantity prioritization principles. Furthermore, an Analytic hierarchy process-based Heterogeneous edge Service dePloyment algorithm for CPSI in IoV, named AHSP, has been designed to determine optimal service deployment strategies. Finally, extensive numerical experimental results demonstrate the effectiveness of AHSP. IEEE

关键词： Computation theory

来源：评论

学校读者我要写书评

暂无评论

Motion compensation and object detection for neuromorphic camera

Motion compensation and object detection for neuromorphic ca...

引用

2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

作者： Wan, Yuxin Fei, Rong Tang, Yu Bai, Xueru Xie, Guo Li, Aimin Xi'An University of Technology School of Computer Science and Engineering Xi'an China Xi'An University of Technology Shaanxi Key Laboratory for Network Computing and Security Technology School of Computer Science and Engineering Xi'an China Xidian University Guangzhou Institute Guangzhou China Xi'An University of Technology School of Computer Science and Engineering Xi'an China Xi'An University of Technology School of Automation and Information Engineering Xi'an China

ISBN: (纸本)9781665401265

Compared to conventional cameras, the new type of vision camera-neuromorphic cameras, which can avoid motion blur and have the advantages of high spatiotemporal resolution, high dynamic range, low latency, etc. In this paper, two motion compensation methods are introduced for operating event flow based on differential neuromorphic camera, one is motion compensation based on event counting image and time image, another is motion compensation based on image of warped events. Comprehensive experimental study is carried on two datasets, the motion compensation method based on event counting image and time image is better at generating clear motion compensation images, highlighting moving objects inconsistent with the background motion, while the motion compensation method based on image of warped events tends to generate motion compensated images of moving objects with more prominent or clearer edges. Simultaneously, two motion compensation methods are further applied to realize the object detection method based on threshold or maximum contrast. The former method can be adapted in many cases, while the latter method can well detect the translational moving target. © 2021 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Fair Streaming Feature Selection

arXiv

引用

arXiv 2024年

作者： Duan, Zhangling Li, Tianci Wu, Xingyu Ling, Zhaolong Yang, Jingye Jia, Zhaohong School of Internet Anhui University Hefei230601 China Department of Computing The Hong Kong Polytechnic University 999077 Hong Kong School of Computer Science and Technology Anhui University Hefei230601 China

Streaming feature selection techniques have become essential in processing real-time data streams, as they facilitate the identification of the most relevant attributes from continuously updating information. Despite their performance, current algorithms to streaming feature selection frequently fall short in managing biases and avoiding discrimination that could be perpetuated by sensitive attributes, potentially leading to unfair outcomes in the resulting models. To address this issue, we propose FairSFS, a novel algorithm for Fair Streaming Feature Selection, to uphold fairness in the feature selection process without compromising the ability to handle data in an online manner. FairSFS adapts to incoming feature vectors by dynamically adjusting the feature set and discerns the correlations between classification attributes and sensitive attributes from this revised set, thereby forestalling the propagation of sensitive data. Empirical evaluations show that FairSFS not only maintains accuracy that is on par with leading streaming feature selection methods and existing fair feature techniques but also significantly improves fairness metrics. © 2024, CC BY.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Editorial:Special Topic on Federated Learning for IoT and Edge computing

引用

ZTE Communications 2022年第3期20卷 1-2页

作者： PAN Yi CUI Laizhong CAI Zhipeng LI Wei Faculty of Computer Science and Control Engineering at Shenzhen Institute of Advanced Technology Chinese Academy of SciencesChina the Academy of the United Nations Sciences and Technology Organization Ukrainian Academy of Engineering Sciences American Institute for Medical and Biological Engineering Institute of Engineering Technology the College of Computer Science and Software Engineering at Shenzhen University China Department of Computer Science and the College of Business at Georgia State University USA the Center for Distributed and High Performance Computing School of Computer ScienceThe University of Sydney the Australia-China Joint Research Center on Energy Informatics and Demand Response Technologies

Recent years have witnessed the proliferation of Internet of Things(IoT),in which billions of devices are connected to the Internet,generating an overwhelming amount of *** is challenging and infeasible to transfer an... 详细信息

关键词： cloud generating data.

来源：评论

学校读者我要写书评

暂无评论

Deep DWT Feature Modeling for Alzheimer’s Disease Prediction: A Unique Approach 1st

Deep DWT Feature Modeling for Alzheimer’s Disease Predictio...

引用

1st International Conference on Machine Learning Algorithms, ICMLA 2024

作者： Tripathy, Santosh Kumar Behera, Chandan Kumar Gadupa, Kartik Shankar Nayak, Rudra Kalyan Department of Computer Science and Engineering National Institute of Technology Patna Bihar Patna India School of Computing Science and Engineering VIT Bhopal University Bhopal-Indore Highway Kothrikalan Madhya Pradesh Sehore466114 India

ISBN: (纸本)9783031758607

The procedure of computer-aided diagnosis for Alzheimer’s disease forecast has gotten better using deep learning principles. Such techniques rely on deep spatial features for classification. Nevertheless, the systems’ capabilities have to be enhanced. Furthermore, deep-frequency domain characteristics for Alzheimer’s diagnosis are understudied. This article fills the said research gaps by exploiting deep frequency features for disease prediction. The model utilizes the high-level and low-level frequency-based DWT features. Then a two-stream deep CNN architecture is designed where the low- and high-level frequency components are inputted. Each stream is a multi-layer of conventional CNN layers. Concatenation of the deep characteristics of the two streams is performed, and then the information is sent to a multi-layer of dense layers for illness classification. Achievement of the suggested model is evaluated by experimenting on OASIS and the Kaggle AD datasets and the obtained results outperform the existing state-of-the-arts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Self-supervised Topic Taxonomy Discovery in the Box Embedding Space

arXiv

引用

arXiv 2024年

作者： Lu, Yuyin Chen, Hegang Mao, Pengbo Rao, Yanghui Xie, Haoran Wang, Fu Lee Li, Qing School of Computer Science and Engineering Sun Yat-sen University Guangzhou China School of Data Science Lingnan University Hong Kong School of Science and Technology Hong Kong Metropolitan University Hong Kong Department of Computing The Hong Kong Polytechnic University Hong Kong

Topic taxonomy discovery aims at uncovering topics of different abstraction levels and constructing hierarchical relations between them. Unfortunately, most of prior work can hardly model semantic scopes of words and topics by holding the Euclidean embedding space assumption. What’s worse, they infer asymmetric hierarchical relations by symmetric distances between topic embeddings. As a result, existing methods suffer from problems of low-quality topics at high abstraction levels and inaccurate hierarchical relations. To alleviate these problems, this paper develops a Box embedding-based Topic Model (BoxTM) that maps words and topics into the box embedding space, where the asymmetric metric is defined to properly infer hierarchical relations among topics. Additionally, our BoxTM explicitly infers upper-level topics based on correlation between specific topics through recursive clustering on topic boxes. Finally, extensive experiments validate high-quality of the topic taxonomy learned by BoxTM. Copyright © 2024, The Authors. All rights reserved.

关键词： Taxonomies

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：