检索结果-内蒙古大学图书馆

An Attention-Based Spatio-temporal Model for Methane Concentration Forecasting in Coal Mine

neural processing LETTERS 2023年第4期55卷 4777-4798页

作者： Gao, Yifei Zhang, Xiaohang Li, Zhengren Du, Yu Beijing Univ Posts & Telecommun Sch Econ & Management Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun Key Lab Trustworthy Distributed Comp & Serv Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun Sch Modern Post Beijing 100876 Peoples R China Beijing Language & Culture Univ Business Sch Beijing 100083 Peoples R China

Methane is a kind of harmful gas produced in the process of coal mining. Because of its flammable and explosive characteristics, it poses a great threat to safety during coal mining. In practice, accurate and real-time gas concentration forecasting is becoming an essential issue for reducing methane risks and accidents. Given the complex spatial correlation and temporal variation of sensor data in a coal mine monitoring system, deep learning algorithms have been widely applied due to their revolutionary feature representation capability. However, existing deep learning models utilize recurrent neural networks, which can barely provide satisfactory accuracy due to their ignorance of realistic working conditions of a coal face or an insufficiency in capturing representative spatio-temporal patterns. In this paper, we propose an attention-based spatio-temporal encoder-decoder network approach, named the ASTED model, for methane concentration forecasting. The ASTED model is built based on the integration of the spatial, temporal and environmental information. Specifically, the multi-attention mechanism is used to learn the dynamic spatio-temporal dependencies, and the feature fusion module is used to incorporate the data from different mine sensors. Finally, we employ the LSTM-based encoder-decoder model to generate the final prediction results. Experiments demonstrate that the ASTED model can obtain the dynamic spatio-temporal correlation from multiple sensor readings and achieve the best performance compared with various state-of-the-art solutions.

关键词： Coal mine safety Spatio-temporal model Attention mechanism Methane concentration forecasting

来源：评论

学校读者我要写书评

暂无评论

Federated Learning Over Coupled Graphs

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2023年第4期34卷 1159-1172页

作者： Lei, Runze Wang, Pinghui Zhao, Junzhou Lan, Lin Tao, Jing Deng, Chao Feng, Junlan Wang, Xidian Guan, Xiaohong XianJiaotong Univ MOE Key Lab Intelligent Networks & Network Secur Xian 710049 Shaanxi Peoples R China China Mobile Res Inst Beijing 100875 Peoples R China China Mobile Grp Design Inst Beijing 100080 Peoples R China Tsinghua Univ Ctr Intelligent & Networked Syst Tsinghua Natl Lab Informat Sci & Technol Beijing 100084 Peoples R China

Graphs are widely used to represent the relations among entities. When one owns the complete data, an entire graph can be easily built, therefore performing analysis on the graph is straightforward. However, in many scenarios, it is impractical to centralize the data due to data privacy concerns. An organization or party only keeps a part of the whole graph data, i.e., graph data is isolated from different parties. Recently, Federated Learning (FL) has been proposed to solve the data isolation issue, mainly for Euclidean data. It is still a challenge to apply FL on graph data because graphs contain topological information which is notorious for its non-IID nature and is hard to partition. In this work, we propose a novel FL framework for graph data, FedCog, to efficiently handle coupled graphs that are a kind of distributed graph data, but widely exist in a variety of real-world applications such as mobile carriers' communication networks and banks' transaction networks. We theoretically prove the correctness and security of FedCog. Experimental results demonstrate that our method FedCog significantly outperforms traditional FL methods on graphs. Remarkably, our FedCog improves the accuracy of node classification tasks by up to 14.7%.

关键词： distributed graph processing federated learning graph neural networks privacy preservation

来源：评论

学校读者我要写书评

暂无评论

Brand Logo Image Classification Based on Deep Convolutional Generative Adversarial network 3

Brand Logo Image Classification Based on Deep Convolutional ...

引用

3rd IEEE International Conference on distributed Computing and Electrical Circuits and Electronics, ICDCECE 2024

作者： Wang, Yang Guangzhou Institute Of Science Technology Guangzhou China

ISBN: (纸本)9798350318609

Brand logo image examines how a critical dimension of logo design, namely the naturalness of the logo color, influences brand design induced by the logo. However, high complexity is observed, requiring higher quality data and susceptibility to overfitting when there is insufficient training data. The implementation of the proposed system involved the utilization of a Deep Convolutional Generative Adversarial network (DCGAN) reduce overfitting and efficiently handle noise and distortion in the logo image, which is essential in brand logo image processing. The brand logo image dataset forwarded to the preprocessing stage included pixel values, which may lead to the loss of contrast, shadow, sharpness, noise removal, and structure in some relevant images. The classification involved DCGAN, aiming to achieve high-quality logo image prediction, noise reduction, and handling of overfitting by incorporating synthetic logo images. The proposed DCGAN method achieves promising performance, including a mean Average Precision (mAP) of 52.25, Accuracy of 97.09%, Precision of 97.01 %, Recall of 98.10%, and F1-score of 97.11 %. This performance is compared to existing methods such as YOLOv3 and Convolutional neural network (CNN) - Long Short-Term Memory (LSTM). © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Exploiting Processor Heterogeneity to Improve Throughput and Reduce Latency for Deep neural network Inference 36

Exploiting Processor Heterogeneity to Improve Throughput and...

引用

36th IEEE International Symposium on Computer Architecture and High-Performance Computing, SBAC-PAD 2024

作者： Beaumont, Olivier David, Jean-Francois Eyraud-Dubois, Lionel Thibault, Samuel Inria Center of the University of Bordeaux Bordeaux France University of Bordeaux Bordeaux France

ISBN: (纸本)9798350356168

The growing popularity of Deep neural networks (DNNs) in a variety of domains, including computer vision, natural language processing, and predictive analytics, has led to an increase in the demand for computing resources. Graphics processing Units (GPUs) are widely used for training and inference of DNNs. However, this exclusive use can quickly lead to saturation of GPU resources while CPU resources remain underutilized. This paper proposes a performance evaluation of a solution that exploits processor heterogeneity by combining the computational power of GPUs and CPUs. A solution is proposed for distributing the computational load across the different processors to optimize their utilization and achieve better performance. A solution for partitioning a DNN model with different computational resources is proposed. This solution transfers part of the load from the GPUs to the CPUs when necessary to reduce latency and increase throughput. The partitioning of DNN models is performed using METIS to balance the computational load to be distributed among the different resources while minimizing communication. The experimental results show that latency and throughput are improved for a number of DNN models. Potential applications include real-time processing systems such as autonomous vehicles, drones, and video surveillance systems where minimizing latency and maximizing throughput are critical. © 2024 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

distributed Device-Specific Anomaly Detection using Deep Feed-Forward neural networks 36

Distributed Device-Specific Anomaly Detection using Deep Fee...

引用

36th IEEE/IFIP network Operations and Management Symposium, NOMS 2023

作者： Lübben, Christian Pahl, Marc-Oliver Technical University of Munich Germany Imt Atlantique France

ISBN: (纸本)9781665477161

The Internet of Things (IoT) requires sophisticated security measures because of heterogeneity and resource constraints. Current approaches in Anomaly Detection (AD) do not meet both challenges. Device-specific AD models can account for the heterogeneity of devices. However, existing approaches fail to run on constrained devices. This paper presents the approach of distributed device-specific AD models. Each model processes only data from one device. Through this simplification of the prediction task, lightweight AD models can be created. They provide the ability to counter the resource constraints of devices. With less requirements in processing power, IoT devices can perform AD on their own. The novel approach improves the optimization metrics detection performance, latency, and model complexity. The evaluation uses the publicly available UNSW-NB15 dataset. It shows that models can be simplified to run on IoT devices. Measurements with a device-specific model on a Raspberry PI show only a little increase in training and processing time compared to central processing on a desktop PC. While the accuracy maintains >98%, the F1-score increases from 0.64 up to 0.89 in the distributed approach. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

An Overview of Different Approaches used for processing Satellite Images

An Overview of Different Approaches used for Processing Sate...

引用

2024 International Conference on distributed Systems, Computer networks and Cybersecurity, ICDSCNC 2024

作者： Chethana, V. Rashmi, S. Visvesveraya Technological University Dayananda Sagar College of Engineering Department of Information Science & Engineering Belagavi India Visvesveraya Technological University Dayananda Sagar College of Engineering Department of Computer Science & Engineering (Data Science) Belagavi India

ISBN: (数字)9798350375442

ISBN: (纸本)9798350375459

In country such as India due to monsoon and frequent climatic changes leads to natural disasters such as floods, drought, landslides, cyclones, forest fire, earthquake and so on. The post disaster badly impacts on human life, economy and environment. As remote sensing technology advanced, the satellite images with high resolution can be processed and analyzed to get the required results. These analyzed results can be used for rescue operations after disaster or to predict any disasters before their occurrences. Some of the applications such as post disaster event management widely use onboard processing to get faster analyzed results. Various operations such as satellite image- enhancement, compression, classification and segmentation are performed on satellite images. Results of these techniques mainly used in applications such as disaster management, Land Use Land Cover, Change detection, landslide and many more. In recent days, satellite image processing are performed using new technical aspects such as deep learning, Artificial intelligence and machine learning to get fast and more accurate results. Edge-Computing-Enabled Deep Learning and Leveraging Deep Learning are the approaches used for satellite image enhancement. Panoptic Segmentation of Multispectral Satellite Images approaches are mainly used for satellite image segmentation. Deep Learning based approach and Genererative Adverserial network architecture are applied to get change detection in satellite images. Lossless algorithm for hyperspectral data compression and neural network lossy compression techniques are preferably used for satellite image compression. Approaches such as Hybrid data model centric approach and Multiscale Convolutional neural network are applied to perform object detection. And also machine learning algorithms are mainly used for classification of satellite images. This paper provides comparative study survey on these various approaches used for processing of satellite images. ©

关键词： convolutional neural network deep learning disaster remote sensing satellite images

来源：评论

学校读者我要写书评

暂无评论

Accuracy evaluation of supervised machine learning classification models for wireless network traffic

引用

INTERNATIONAL JOURNAL OF COMMUNICATION networkS AND distributed SYSTEMS 2022年第6期28卷 655-678页

作者： Grabs, Elans Petersons, Ernests Efrosinin, Dmitry Ipatovs, Aleksandrs Kluga, Janis Sturm, Valentin Riga Tech Univ Telecommun Inst Azenes St LV-1048 Riga Latvia Johannes Kepler Univ Linz Inst Stochast Altenberger Str 69 A-4040 Linz Austria

The article contains results of training and testing machine learning models with captured network traffic data. The main goal is to perform classification of video traffic in computer networks. Multiple performance metrics have been evaluated for commonly used classic supervised machine learning algorithms, as well as more advanced convolutional neural network model (for comparison). The article describes in detail the experimental setup, traffic pre-processing procedure, features extraction with different traffic window length and model parameters for training/testing. The article provides some experimental results in the form of tables and 3D surface plots. The conclusion of the article summarises the main findings and outlines the future study directions.

关键词： accuracy classification models features extraction network traffic performance metrics statistical parameters supervised machine learning traffic intensity window length wireless networks convolutional neural network CNN

来源：评论

学校读者我要写书评

暂无评论

Enhancing Small Object Detection in Aerial Imagery Based on Strong Feature Extraction and Batch Dimension

Enhancing Small Object Detection in Aerial Imagery Based on ...

引用

International Joint Conference on neural networks (IJCNN)

作者： Cao, Bingnan Wang, Di Guo, Ying Zhang, Hu Qilu Univ Technol Shandong Comp Sci Ctr Natl Supercomp Ctr Jinan Key Lab Comp Power Network & Informat Secur Minist EducShandong Acad Sci Jinan Peoples R China Shandong Fundamental Res Ctr Comp Sci Shandong Prov Key Lab Comp Networks Jinan Peoples R China City Univ Macau Inst Data Sci Macau Peoples R China

ISBN: (纸本)9798350359329;9798350359312

Object detection is a fundamental component of computer vision, playing a pivotal role in various applications. However, the accurate detection of small and densely distributed objects remains a challenging problem in this field. This challenge is particularly exacerbated in the context of aerial imagery, characterized by its distinctive bird's-eye view, intricate backgrounds, and the variability in object appearances. This paper addresses these persistent challenges in object detection, with a focus on the specific difficulties posed by aerial images. We propose a deformable end-to-end object detection with transformers (DETR)-based framework to enhance small object detection accuracy, ultimately contributing to improved computer vision capabilities in domains like remote sensing, surveillance, and autonomous aerial systems. Firstly, in order to aggregate the entire input sequence information in the backbone network and improve the detection accuracy of small objects, we propose DMCA based on deformable features and attention mechanisms. Secondly, in order to capture and model the relationships between samples for dense pixel-level representations in small objects and improve the detection accuracy of small objects, we try to introduce BatchEncoder by implementing an encoder in the batch dimension. Experimental results show that, compared to the baseline, our method significantly improves the accuracy of small object detection in aerial images. The processing and analysis of a large amount of remote sensing data require powerful computing power. The computing power Internet can integrate the scattered computing resources to provide flexible and scalable computing power to meet the computing needs of different sizes and types.

关键词： small object detection remote sensing DMCA(Deformable Multiscale Channel Attention) BatchEncoder

来源：评论

学校读者我要写书评

暂无评论

Dystri: A Dynamic Inference based distributed DNN Service Framework on Edge 23

Dystri: A Dynamic Inference based Distributed DNN Service Fr...

引用

52nd International Conference on Parallel processing (ICPP)

作者： Hou, Xueyu Guan, Yongjie Han, Tao New Jersey Inst Technol Newark NJ 07102 USA

ISBN: (纸本)9798400708435

Deep neural network (DNN) inference poses unique challenges in serving computational requests due to high request intensity, concurrent multi-user scenarios, and diverse heterogeneous service types. Simultaneously, mobile and edge devices provide users with enhanced computational capabilities, enabling them to utilize local resources for deep inference processing. Moreover, dynamic inference techniques allow content-based computational cost selection per request. This paper presents Dystri, an innovative framework devised to facilitate dynamic inference on distributed edge infrastructure, thereby accommodating multiple heterogeneous users. Dystri offers a broad applicability in practical environments, en-compassing heterogeneous device types, DNN-based applications, and dynamic inference techniques, surpassing the state-of-the-art (SOTA) approaches. With distributed controllers and a global coordinator, Dystri allows per-request, per-user adjustments of qualityof-service, ensuring instantaneous, flexible, and discrete control. The decoupled workflows in Dystri naturally support user heterogeneity and scalability, addressing crucial aspects overlooked by existing SOTA works. Our evaluation involves three multi-user, heterogeneous DNN inference service platforms deployed on distributed edge infrastructure, encompassing seven DNN applications. Results show Dystri achieves near-zero deadline misses and excels in adapting to varying user numbers and request intensities. Dystri outperforms baselines with accuracy improvement up to 95x.

关键词： MLaaS dynamic inference edge computing

来源：评论

学校读者我要写书评

暂无评论

Pavlovian Conditioning Modeling Using Wireless Spiking neural network 20

Pavlovian Conditioning Modeling Using Wireless Spiking Neura...

引用

20th International SoC Design Conference (ISOCC)

作者： Lin, Jiaying Nagazawa, Ryuji Kien Nguyen Sekiya, Hiroo Torikai, Hiroyuki Hasegawa, Mikio Hwang, Won-Joo Chiba Univ Grad Sch Sci & Engn Chiba Chiba 2638522 Japan Hosei Univ Grad Sch Sci & Engn Koganei Tokyo 1848584 Japan Tokyo Univ Sci Fac Engn Grad Sch Sinjuku Tokyo 1258585 Japan Pusan Natl Univ Dept Informat Convergence Engn Artificial Intelli Busan 46241 South Korea

ISBN: (纸本)9798350327038

To address the communication congestion and high energy consumption issues in IoT networks, prior research has proposed a distributed information processing system called WiBIC. As an exploratory step towards implementing the basic functionalities of the WiBIC system, this study focuses on the wireless integration of SNNs. Considering the data transmission and learning characteristics of SNNs, we propose using the APCMA protocol to effectively stabilize the operation of the wireless SNN. By modeling Pavlovian conditioning on this network, the learning capabilities of the implemented system can be determined.

关键词： Spiking neural network wireless communication

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：