检索结果-内蒙古大学图书馆

T³Planner: Multi-Phase Planning Across Structure-Constrained Optical, IP, and Routing Topologies

IEEE Journal on Selected Areas in Communications 2025年第5期43卷 1823-1839页

作者： Yijun Hao Shusen Yang Fang Li Yifan Zhang Cong Zhao Xuebin Ren Peng Zhao Chenren Xu Shibo Wang National Engineering Laboratory for Big Data Analytics Xi’an Jiaotong University Xi’an China National Engineering Laboratory for Big Data Analytics and the Ministry of Education Key Laboratory for Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an China School of Computer Science Peking University Beijing China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong China National Engineering Laboratory for Big Data Analytics XiÃan Jiaotong University XiÃan China

Network topology planning is an essential multi-phase process to build and jointly optimize the multi-layer network topologies in wide-area networks (WANs). Most existing practices target single-phase/layer planning, and are incapable of satisfying all rigorous topological structure constraints (e.g., dual-homing rings) defined by network standards and operators, especially in large-scale networks. These significantly limit their usability and performance in production networks. We consider a general topology planning problem with typical structure constraints over three essential phases (greenfield, reconfiguration, and site expansion) and topological layers (optical, IP, and routing topologies). We present, T3Planner, a novel practical solver to this problem in production. Specifically, we develop a structure-driven encoder based on graph neural network (GNN) for concise structure encoding, and design a new learning framework with optical-centric layer compression/reconstruction and rule-aided reinforcement learning (RL) for fast convergence and high performance. Extensive experiments on nine real topologies demonstrate that T3Planner scales to large optical networks with hundreds of sites, saves 46.6% cost, and supports $3.12\times $ more demand when compared to related existing approaches.

关键词： Topology Network topology Planning IP networks Optical fiber networks Routing Optical fibers Optical feedback Production Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Monocular Camera-Based Visual Truck Type Recognition

SSRN

引用

SSRN 2024年

作者： Luo, Haibo Ye, Huaisheng Feng, Chao Shi, Xin Xie, Guoqi School of Computer and Data Science Minjiang University Fuzhou350018 China College of Computer and Data Science Fuzhou University Fuzhou350018 China College of Network and Information Security Fujian Agriculture and Forestry University Fuzhou350018 China Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University Changsha410082 China College of Computer Science and Electronic Engineering Hunan University Changsha410082 China

Truck overload and over-limit are the primary causes of infrastructure damage and traffic safety accidents. In the past 2 years, researchers have started to deploy intelligent Internet of Things system at the source of truck delivery to enforce permit control. Truck type recognition is the core of such a system. However, existing methods suffer from limitations such as low recognition accuracy, high costs, difficult maintenance, and the need for road surface modifications. This study proposes a monocular camera-based visual truck type recognition method. First, the optimized lightweight deep neural network based on machine vision is utilized to detect truckaxles and recognize wheel hub categories. Next, a tracking algorithm and the video streamcontext information are employed to complete the robust statistics of the axle by neighborhoodcounting. Finally, the proposed method further achieves truck type recognition by leveragingavailable prior knowledge. Experimental results demonstrate that this method exhibits superiortime efficiency compared with existing algorithms, and it reduces average processing time by92.37% to 96.24% while ensuring high accuracy. Moreover, the method has been deployed onembedded devices with low computing power to enable real-time operation onsite. © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

MeHyper: Accelerating Hypergraph Neural Networks by Exploring Implicit dataflows

MeHyper: Accelerating Hypergraph Neural Networks by Explorin...

引用

IEEE Symposium on High-Performance computer Architecture

作者： Wenju Zhao Pengcheng Yao Dan Chen Long Zheng Xiaofei Liao Qinggang Wang Shaobo Ma Yu Li Haifeng Liu Wenjing Xiao Yufei Sun Bing Zhu Hai Jin Jingling Xue National Engineering Research Center for Big Data Technology and System/Services Computing Technology and System Lab/Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China School of Computing National University of Singapore Singapore School of Computer Electronics and Information Guangxi University NanNing China School of Computer Science and Engineering University of New South Wales Sydney NSW Australia

ISBN: (数字)9798331506476

ISBN: (纸本)9798331506483

Hypergraph Neural Networks (HGNNs) are increasingly utilized to analyze complex inter-entity relationships. Traditional HGNN systems, based on a hyperedge-centric dataflow model, independently process aggregation tasks for hyperedges and vertices, leading to significant computational redundancy. This redundancy arises from recalculating shared information across different tasks. For the first time, we identify and harness implicit dataflows (i.e., dependencies) within HGNNs, introducing the microedge concept to effectively capture and reuse intricate shared information among aggregation tasks, thereby minimizing redundant computations. We have developed a new microedge-centric dataflow model that processes shared information as fine-grained microedge aggregation tasks. This dataflow model is supported by the Read-Process-Activate-Generate execution model, which aims to optimize parallelism among these tasks. Furthermore, our newly developed MeHyper, a microedge-centric HGNN accelerator, incorporates a decoupled pipeline for improved computational parallelism and a hierarchical feature management strategy to reduce off-chip memory accesses for large volumes of intermediate feature vectors generated. Our evaluation demonstrates that MeHyper substantially outperforms the leading CPUbased system PyG-CPU and the GPU-based system HyperGef, delivering performance improvements of $1,032.23 \times$ and $10.51 \times$, and energy efficiencies of $1,169.03 \times$ and $9.96 \times$, respectively.

关键词： Computational modeling Neural networks Redundancy Pipelines Memory management Parallel processing Vectors Energy efficiency Hardware acceleration

来源：评论

学校读者我要写书评

暂无评论

Low-complexity Multi-Channel Speaker Extraction with Pure Speech Cues

Low-complexity Multi-Channel Speaker Extraction with Pure Sp...

引用

Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

作者： Bang Zeng Hongbin Suo Yulong Wan Ming Li Data Science Research Center Duke Kunshan University Kunshan China School of Computer Science Wuhan University Wuhan China Data&AI Engineering System OPPO Beijing China

Most multi-channel speaker extraction schemes use the target speaker’s location information as a reference, which must be known in advance or derived from visual cues. In addition, memory and computation costs are enormous when the model deals with the fusion input. In this paper, we propose Speaker-extraction-and-filter Network (SeafNet), which is a low-complexity multi-channel speaker extraction network with only speech cues. Specifically, the SeafNet separates the mixture by utilizing the correlation between an estimation of target speaker on reference channel and the mixed input on rest channels. Experimental results show that compared with the baseline, the SeafNet model achieves 6.4% relative SISNRi improvement on the fixed geometry array and 8.9% average relative SISNRi improvement on the ad-hoc array. Meanwhile, the SeafNet achieves 60% relative reduction in the number of parameters and 42% relative reduction in the computational cost.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Object Detection Method for Bank Operation and Maintenance Scenarios 5

A Lightweight Object Detection Method for Bank Operation and...

引用

5th International Conference on Pattern Recognition and Artificial Intelligence, PRAI 2022

作者： Xue, Junmin Rao, Yunbo Wu, Jiangang Gou, Haosong Liu, Yong Yang, Yang University of Electronic Science and Technology of China School of Information and Software Engineering Chengdu China Data Center of Postal Savings Bank of China Beijing China Chengdu Aerospace Science and Industry Big Data Research Institute Corporation Limited Chengdu China Sichuan Institute of Computer Sciences Chengdu China

ISBN: (数字)9781665499163

ISBN: (纸本)9781665499163

Due to the complex background of bank operation and maintenance scenarios, it is extremely difficult to detect and analyze the object of operation and maintenance scenarios. The traditional target detection algorithms, such as F-RCNN, are not suitable for object detection. We propose a lightweight object detection method for bank operation and maintenance scenarios based on deep learning. The proposed method highlights two improvements based on Yolov5. The first adjustment of it is this method constructs a new convolution method and a new activation function. Also, we use the multi-scale output strategy of Feature Pyramid Networks (FPN) to develop a new lightweight feature extraction algorithm. These changes we made are based on the real-life situation, such as bank operation and maintenance scenarios. Experiments have been carried out on the same data set with multiple object detection algorithms, and the experimental results show that the proposed method reduces the complexity of the activation function. Meanwhile, the module is flexible, and can be easily applied to other similar business scenario detection models. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Service-Based Event Penetration from IoT Sensors to Businesses: a Case Study

Service-Based Event Penetration from IoT Sensors to Business...

引用

2022 CCF International Conference on Service science, CCF ICSS 2022

作者： Wang, Guiling Fang, Jun Wang, Jing Yu, Jian Zhang, Liang Han, Yanbo Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data Beijing China North China University of Technology School of Information Science and Technology Beijing China Auckland University of Technology Department of Computer Science Auckland New Zealand Shanghai Key Laboratory of Data Science Shanghai China Shanghai Institute of Intelligent Electronics & Systems Shanghai China

ISBN: (数字)9781665498616

ISBN: (纸本)9781665498616

By leveraging IoT Big data, BPM can gain real-time physical world information to make faster and more accurate decisions, but there is a technical gap between IoT sensors and businesses. To bridge the gap, an event penetration mechanism from IoT sensors to business processes is proposed along a practical case study. This paper presents a concrete IoT-BPM application case dealing with seaborne safety ensurance in transporting liquefied natural gas (LNG), analyzes its technical challenges, and examines the feasibility and supposed effects of the BRIBOT approach. © 2022 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Elevating Cystic Fibrosis Detection in Lungs using HRCT Images with a Cutting-Edge CNN-Based Approach

Elevating Cystic Fibrosis Detection in Lungs using HRCT Imag...

引用

Advanced Computing & Communication Technologies (ICACCTech), International Conference on

作者： Amrutha Sriya Ganjam Kishor Kumar Reddy C D Radhika Karthik Ganjam Monika Singh.T Keerthi Sai Akuthota Department of Computer Science Stanley College of engineering and technology for womenabids Hyderabad India Department School of informatics MSc Data Science University of Edinburgh Edinburgh Scotland UK Deptartment. of Computer Science and Engineering Stanley College of Engineering and Technology for Women Hyderabad India

ISBN: (数字)9798331519056

ISBN: (纸本)9798331519063

The creation of an algorithm for recognizing pathological abnormalities in cystic fibrosis is investigated in this paper using the CNN model with a modified psp-net. Currently, Decision Trees, Random Forests, PSP Nets, and Neural Networks are utilized in the diagnosis of cystic fibrosis. Since convolutional neural networks (CNNs) can process complicated picture data rapidly and efficiently, the goal of this study is to use CNNs for the detection of anomalies associated with cystic fibrosis. The method groups distinct annotated images into a simple and efficient structure, runs the set of images through a multiscale CNN procedure, and precisely locates the lung region affected by cystic fibrosis. The result of this paper demonstrated that differences in the training dataset can impact performance, but annotating CT images and categorizing them in terms of similar pathologies can improve the accuracy of the model. The proposed CNN model achieved an Accuracy of 84 %, Precision of 74%, Recall of 79%, F1-Score of 72%, error rate of 16% Which is better when compared with existing approaches.

关键词： Training Pathology Accuracy Image recognition Error analysis Computed tomography Lungs data models Convolutional neural networks Diseases

来源：评论

学校读者我要写书评

暂无评论

Calibrating a Deep Neural Network with Its Predecessors

arXiv

引用

arXiv 2023年

作者： Tao, Linwei Dong, Minjing Liu, Daochang Sun, Changming Xu, Chang School of Computer Science Faculty of Engineering University of Sydney Australia CSIRO's Data61 Australia

Confidence calibration - the process to calibrate the output probability distribution of neural networks - is essential for safety-critical applications of such networks. Recent works verify the link between mis-calibration and overfitting. However, early stopping, as a well-known technique to mitigate overfitting, fails to calibrate networks. In this work, we study the limitions of early stopping and comprehensively analyze the overfitting problem of a network considering each individual block. We then propose a novel regularization method, predecessor combination search (PCS), to improve calibration by searching a combination of best-fitting block predecessors, where block predecessors are the corresponding network blocks with weight parameters from earlier training stages. PCS achieves the state-of-the-art calibration performance on multiple datasets and architectures. In addition, PCS improves model robustness under dataset distribution shift. Supplementary material and code are available at https://***/Linwei94/PCS. © 2023, CC BY.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Unbiased Approximations for Stationary Distributions of McKean-Vlasov SDEs

arXiv

引用

arXiv 2024年

作者： Awadelkarim, Elsiddig Chada, Neil K. Jasra, Ajay Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Department of Mathematics City University of Hong Kong Hong Kong School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen China

We consider the development of unbiased estimators, to approximate the stationary distribution of Mckean-Vlasov stochastic differential equations (MVSDEs). These are an important class of processes, which frequently appear in applications such as mathematical finance, biology and opinion dynamics. Typically the stationary distribution is unknown and indeed one cannot simulate such processes exactly. As a result one commonly requires a time-discretization scheme which results in a discretization bias and a bias from not being able to simulate the associated stationary distribution. To overcome this bias, we present a new unbiased estimator taking motivation from the literature on unbiased Monte Carlo. We prove the unbiasedness of our estimator, under assumptions. In order to prove this we require developing ergodicity results of various discrete time processes, through an appropriate discretization scheme, towards the invariant measure. Numerous numerical experiments are provided, on a range of MVSDEs, to demonstrate the effectiveness of our unbiased estimator. Such examples include the Currie-Weiss model, a 3D neuroscience model and a parameter estimation problem. © 2024, CC BY-NC-SA.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Fedward: Flexible Federated Backdoor Defense Framework with Non-IID data

Fedward: Flexible Federated Backdoor Defense Framework with ...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Zekai Chen Fuyi Wang Zhiwei Zheng Ximeng Liu Yujie Lin College of Computer Science and Big Data Fuzhou University Fuzhou China School of Information Technology Deakin University

Federated learning (FL) enables multiple clients to collaboratively train deep learning models while considering sensitive local datasets’ privacy. However, adversaries can manipulate datasets and upload models by injecting triggers for federated backdoor attacks (FBA). Existing defense strategies against FBA consider specific and limited attacker models, and a sufficient amount of noise to be injected only mitigates rather than eliminates FBA. To address these deficiencies, we introduce a Flexible Federated Backdoor Defense Framework (Fedward) to ensure the elimination of adversarial backdoors. We decompose FBA into various attacks, and design amplified magnitude sparsification (AmGrad) and adaptive OPTICS clustering (AutoOPTICS) to address each attack. Meanwhile, Fedward uses the adaptive clipping method by regarding the number of samples in the benign group as constraints on the boundary. This ensures that Fedward can maintain the performance for the Non-IID scenario. We conduct experimental evaluations over three benchmark datasets and thoroughly compare them to state-of-the-art studies. The results demonstrate the promising defense performance from Fedward, moderately improved by 33% ∼ 75% in clustering defense methods, and 96.98%, 90.74%, and 89.8% for Non-IID to the utmost extent for the average FBA success rate over MNIST, FMNIST, and CIFAR10, respectively.

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：