检索结果-内蒙古大学图书馆

P-SCADA - A novel area and energy efficient FPGA architectures for LSTM prediction of heart arrthymias in biot applications

引用

EXPERT SYSTEMS 2022年第3期39卷

作者： Varadharajan, Senthil Kumaran Nallasamy, Viswanathan Mahendra Engn Coll Autonomous Dept Elect & Commun Engn Namakkal Tamil Nadu India

Recurrent neural networks (RNN) are extensively used to determine the optimal solutions to the various class recognition problems such as image processing, prediction of biomedical data and speech recognition. With the gradient problems, RNN is slowing losing its shade which is replaced by the Long short term memory (LSTM). However the hardware implementation of the LSTM requires more challenge due to its complexity and high power consumption which makes it unsuitable for implementing in Biological Internet of things networks for prediction of heart diseases. Several algorithms were proposed for an effective implementation of LSTM, but hand-offs between the performance and utilization still needs improvisation. The paper proposes the novel energy efficient and high performance architecture Pipelined Stochastic Adaptive distributed Architectures (P-SCADA) for LSTM networks. In this architecture, hybrid structure has been developed with the help of new distributed arithmetic stochastic computing (DSC) along with the binary circuits to advance the performance of the FPGA such as energy, area and accuracy. The proposed system has been implemented in ARTIX-7 FPGA with special purpose software has been designed and evaluated with different ECG datasets. For the different series data, area utilization is about 40%-44% and power consumption is about 20%-25% with the prediction of accuracy of 98%. Moreover the proposed architecture has been compared with the other existing architecture such as SPARSE architectures, normal stochastic architectures in which the proposed architecture excels in terms area, power and efficiency.

关键词： long short term memory (LSTM) recurrent neural network (RNN) SPARSE architectures stochastic computing

来源：评论

学校读者我要写书评

暂无评论

EF-BV: A Unified Theory of Error Feedback and Variance Reduction Mechanisms for Biased and Unbiased Compression in distributed Optimization 36

EF-BV: A Unified Theory of Error Feedback and Variance Reduc...

引用

36th Conference on neural Information processing Systems (NeurIPS)

作者： Condat, Laurent Yi, Kai Richtarik, Peter King Abdullah Univ Sci & Technol KAUST Thuwal 239556900 Saudi Arabia

ISBN: (纸本)9781713871088

In distributed or federated optimization and learning, communication between the different computing units is often the bottleneck and gradient compression is widely used to reduce the number of bits sent within each communication round of iterative methods. There are two classes of compression operators and separate algorithms making use of them. In the case of unbiased random compressors with bounded variance (e.g., rand-k), the DIANA algorithm of Mishchenko et al. (2019), which implements a variance reduction technique for handling the variance introduced by compression, is the current state of the art. In the case of biased and contractive compressors (e.g., top-k), the EF21 algorithm of Richtarik et al. (2021), which instead implements an error-feedback mechanism, is the current state of the art. These two classes of compression schemes and algorithms are distinct, with different analyses and proof techniques. In this paper, we unify them into a single framework and propose a new algorithm, recovering DIANA and EF21 as particular cases. Our general approach works with a new, larger class of compressors, which has two parameters, the bias and the variance, and includes unbiased and biased compressors as particular cases. This allows us to inherit the best of the two worlds: like EF21 and unlike DIANA, biased compressors, like top-k, whose good performance in practice is recognized, can be used. And like DIANA and unlike EF21, independent randomness at the compressors allows to mitigate the effects of compression, with the convergence rate improving when the number of parallel workers is large. This is the first time that an algorithm with all these features is proposed. We prove its linear convergence under certain conditions. Our approach takes a step towards better understanding of two so-far distinct worlds of communication-efficient distributed learning.

关键词： Compressors

来源：评论

学校读者我要写书评

暂无评论

Performance Modeling for distributed Training of Convolutional neural networks 29

Performance Modeling for Distributed Training of Convolution...

引用

29th Euromicro International Conference on Parallel, distributed and network-Based processing (PDP)

作者： Castello, Adrian Catalan, Mar Dolz, Manuel F. Mestre, Jose, I Quintana-Orti, Enrique S. Duato, Jose Univ Jaume 1 Castellon de La Plana Spain Univ Politecn Valencia Valencia Spain

ISBN: (纸本)9781665414555

We perform a theoretical analysis comparing the scalability of data versus model parallelism, applied to the distributed training of deep convolutional neural networks (CNNs), along live axes: batch size, node (floating-point) arithmetic performance, node memory bandwidth, network link bandwidth, and cluster dimension. Our study relies on analytical performance models that can he configured to reproduce the components and organization of the CNN model as well as the hardware configuration of the target distributed platform. In addition, we provide evidence of the accuracy of the analytical models by performing a validation against a Python library for distributed deep learning training.

关键词： Deep neural networks (DNNs) distributed training analytical modeling clusters

来源：评论

学校读者我要写书评

暂无评论

Multi-scale Spin Convolutional neural network for Typhoon Intensity Prediction 29

Multi-scale Spin Convolutional Neural Network for Typhoon In...

引用

29th IEEE International Conference on Parallel and distributed Systems, ICPADS 2023

作者： Li, Longjie Zhu, Yanmin Zhang, Han Jing, Mengyuan Zhang, Haoyu Wang, Zhaobo Zang, Tianzi Shanghai Jiao Tong University China The Second Institute of Oceanography of China China NanJing University of Aeronautics and Astronautics China

ISBN: (纸本)9798350330717

Typhoons, formidable natural phenomena, typically unleash a trail of destruction, inflicting severe wind damage, floods, and even triggering tsunamis in coastal regions. Therefore, accurate prediction of typhoon intensity carries immense significance, both in theory and practical applications. Convolutional neural networks (CNNs) have shown superior capability in modelling spatial and temporal data. However, conventional convolutional operators are very sensitive to rotations, and is not good at modelling hierarchical structures. Since typhoon intensity data are typical spatiotemporal data that possess a distinctive rotational structure and a multi-layered hierarchy, conventional CNN based models fall short in capturing high-order spatial information, thereby limiting the accuracy of predictions. To address the limitations of conventional convolutional operators, we introduce a typhoon intensity prediction model called Multi-scale Spin Convolutional neural network (MS-CNN). Our model rotates the convolution kernel approximately to better model the rotating horizontal structure of the typhoons. Additionally, it employs a fine-grained multi-scale architecture to capture the hierarchical structure of typhoons. We conducted a series of comprehensive experiments using typhoon track dataset in the Western North Pacific. The results demonstrate that the superior performance of our MS-CNN model compared to other state-of-the-art models, marking a substantial advancement in typhoon intensity prediction tasks. © 2023 IEEE.

关键词： Convolutional neural network Spatial-temporal Prediction Typhoon Intensity Prediction

来源：评论

学校读者我要写书评

暂无评论

User Mobility Prediction in 5G networks Using Recurrent neural networks

User Mobility Prediction in 5G Networks Using Recurrent Neur...

引用

2024 IEEE International Conference on Communication, Computing and Signal processing, IICCCS 2024

作者： Shrestha, Anubhav Sharma, Vikalp Hussein, Layth Aishwarya, M. Satyanarayana, A.N. Saimanohar, T. Chandigarh Group of Colleges Chandigarh Engineering College Department of Computer Application Punjab Jhanjeri Mohali140307 India Ies College of Technology Department of Computer Science & Engineering M.P. Bhopal India The Islamic University College Of Technical Engineering Department Of Computers Techniques Engineering Najaf Iraq The Islamic University Of Al Diwaniyah College Of Technical Engineering Department Of Computers Techniques Engineering Al Diwaniyah Iraq Prince Shri Venkateshwara Padmavathy Engineering College Chennai India Sreenidhi Institute of Science and Technology Ece Department Hyderabad India Saveetha Institute of Medical and Technical Sciences Department CSE-AI Tamilnadu Chennai India

ISBN: (纸本)9798350390759

The use of new possibilities introduced by 5G networks also creates new problems and concerns, specifically in the field of user mobility in wireless communication systems. In this paper, the Authors investigate the effectiveness of Recurrent neural networks (RNNs) in the mobility prediction context in order to exploit the temporal component in the users movement data for improved prediction performance. To drive this process, the input datasets are obtained from the GeoLife GPS Trajectories dataset, then follow the data pre-processing steps, constructing an LSTM-based RNN model, and using Python with TensorFlow and Keras. From the analyses, we detected that the accomplishment of using LSTM and hybrid RNN-CNN is significant and higher than the traditional approaches of EN, RFM, and CL. The test results demonstrate that the LSTM model can reduce the MAE and RMSE values by 19. 49% and 14. 82%, respectively. However, the execution of these models has some issues that include compute time, instantaneous prediction that is needed in many real-world applications, and case. Such problems can be solved with the help of such approaches as edge computing and distributed processing. Besides, it is pivotal to put measures in place to maintain the confidentiality and security of user information. This paper highlights the applicative potential of RNNs in redesigning user mobility prediction in 5G networks, and thus helping toward a better ownership of the network players and assets. © 2024 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Auto-GNAS: A Parallel Graph neural Architecture Search Framework

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2022年第11期33卷 3117-3128页

作者： Chen, Jiamin Gao, Jianliang Chen, Yibo Oloulade, Babatounde Moctard Lyu, Tengfei Li, Zhao Cent South Univ Sch Comp Sci & Engn Changsha 410083 Hunan Peoples R China State Grid Hunan Elect Power Co Ltd Informat & Commun Branch Changsha 410007 Peoples R China Alibaba Zhejiang Univ Joint Res Inst Frontier Tec Ecommerce Ranking & Recommendat Syst Hangzhou 311121 Zhejiang Peoples R China

Graph neural networks (GNNs) have received much attention as GNNs have recently been successfully applied on non-euclidean data. However, artificially designed graph neural networks often fail to get satisfactory model performance for a given graph data. Graph neural architecture search effectively constructs the GNNs that achieve the expected model performance with the rise of automatic machine learning. The challenge is efficiently and automatically getting the optimal GNN architecture in a vast search space. Existing search methods serially evaluate the GNN architectures, severely limiting system efficiency. To solve these problems, we develop an Automatic Graph neural Architecture Search framework (Auto-GNAS) with parallel estimation to implement an automatic graph neural search process that requires almost no manual intervention. In Auto-GNAS, we design the search algorithm with multiple genetic searchers. Each searcher can simultaneously use evaluation feedback information, information entropy, and search results from other searchers based on sharing mechanism to improve the search efficiency. As far as we know, this is the first work using parallel computing to improve the system efficiency of graph neural architecture search. According to the experiment on the real datasets, Auto-GNAS obtain competitive model performance and better search efficiency than other search algorithms. Since the parallel estimation ability of Auto-GNAS is independent of search algorithms, we expand different search algorithms based on Auto-GNAS for scalability experiments. The results show that Auto-GNAS with varying search algorithms can achieve nearly linear acceleration with the increase of computing resources.

关键词： Computer architecture Estimation Graph neural networks Genetics Search problems Prediction algorithms Parallel processing neural architecture search parallel search graph neural network

来源：评论

学校读者我要写书评

暂无评论

Efficient distributed Parallel Inference Strategies via Block-based DNN Structure in Edge-to-IoT Continuum 24

Efficient Distributed Parallel Inference Strategies via Bloc...

引用

9th International Conference on Intelligent Information Technology (ICIIT)

作者： Choi, Inhun Akhter, Sharmen Jeong, Hong-Ju Huh, Eui-Nam Kyung Hee Univ Dept Comp Sci & Engn Gyeonggi Do 17104 South Korea Kyung Hee Univ Dept Artificial Intelligence Yeonggi Do 17104 South Korea

ISBN: (纸本)9798400716713

Recently, AI and deep neural networks have found extensive applications in mobile devices, drones, carts, and more. To meet the demands of processing large-scale data and providing DNN inference services with minimal latency, there is a need. However, IoT devices, with their limited computing capabilities, are not well-suited for AI inference. Moreover, considering the diverse requirements of different services, it is necessary to provide inference services that address these variations. To address these challenges, many previous studies have explored collaborative approaches between edge servers and cloud servers by partitioning DNN models. However, these methods face difficulties in finding optimal partitioning points for splitting DNN models and are heavily influenced by network bandwidth since intermediate computation results need to be transmitted to other devices. In this paper, we propose the Adaptive block-based DNN network inference framework. This involves breaking down a large DNN model into block-level networks, training them using knowledge distillation techniques to enable inference only through each block network. Subsequently, dynamic block-level inference calculations are offloaded based on the computing capabilities of edge clusters, providing inference results. Even when using multiple devices, our method is not affected by network bandwidth since only input images need to be transmitted. Experimental results demonstrate that our approach consistently reduces inference latency as the number of devices increases. Additionally, by controlling the trade-off between latency and accuracy, we can provide inference services tailored to various latency requirements.

关键词： DNN inference Block-based DNN network trade-off between latency and accuracy

来源：评论

学校读者我要写书评

暂无评论

Synchronization of Complex-Valued SICNNs with distributed Delays via a Module-phase-type Controller 10th

Synchronization of Complex-Valued SICNNs with Distributed De...

引用

10th International Conference on Communications, Signal processing, and Systems

作者： Chen, Qiuyuan Bin, Honghua Huang, Zhenkun Chen, Chao Jimei Univ Sch Sci Xiamen 361021 Peoples R China

ISBN: (纸本)9789811903908;9789811903892

This paper focuses on synchronization for complex-valued shunting inhibitory cellular neural networks (SICNNs) with distributed delays and designs a novel feedback controller to ensure module-phase synchronization. For the discussion of module-phase synchronization, a lemma is given to show the existence of the bounded solution of the drive system. By constructing a Lyapunov functional and employing the inequality technique, sufficient conditions for module-phase synchronization of complex-valued SICNNs are derived. Finally, the validity of obtained results is demonstrated by a numerical example.

关键词： Module-phase synchronization Complex-valued SICNNs distributed delays Feedback control

来源：评论

学校读者我要写书评

暂无评论

Classification of Remote Sensing Images based on distributed Convolutional neural network Model

引用

International Journal of Performability Engineering 2019年第6期15卷 1508-1517页

作者： Chen, Guanyu Cai, Zhihua Li, Xiang School of Computer Science China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Intelligent Geo-Information Processing China University of Geosciences Wuhan430074 China

With the network model architecture of Google Inception, research is conducted on issues such as the structural design of the model, data preprocessing, tuning of training parameters, computing clusters in a distributed environment, and multi-machine parallel training. According to the performances of different deep neural network models on different data sets, the Google Inception V3 depth network model is used as the prototype to conduct the tuning of training parameters, and the classification of remote sensing images is then realized with this model in the single-machine environment. Furthermore, due to the effectiveness of distributed systems for very large data sets and compute-intensive applications, a data parallel training scheme based on the distributed platform is designed for the convolution neural network model with more complex data form, larger quantity of parameters, and more network levels, after studying the mainstream designs of the distributed machine learning and analyzing the training methods and steps of the convolutional neural network model in a multi-machine environment. It greatly improves the training time of the model, and then the classification of remote sensing images under distributed clusters is realized. © 2019 Totem Publisher, Inc. All rights reserved.

关键词： neural network models

来源：评论

学校读者我要写书评

暂无评论

VIDEO ANALYTIC DATA REDUCTION MODEL FOR FOG COMPUTING

VIDEO ANALYTIC DATA REDUCTION MODEL FOR FOG COMPUTING

引用

70th Annual Modeling and Simulation Conference (ANNSIM)

作者： Abhari, Abdolreza Pudasaini, Dipak Ryerson Univ Distributed Syst & Multimedia Proc Lab DSMP Lab Dept Comp Sci 350 Victoria St Toronto ON Canada

ISBN: (数字)9781713852889

ISBN: (纸本)9781713852889

The large volumes of video data cause network congestion and high latency in the centralized cloud computing system. Fog computing architecture that enables employing edge devices has already been used to address these problems. This paper proposes an application model called Video Analytic Data Reduction Model (VADRM) that divides video analytic jobs into smaller tasks with fewer processing requirements. The prototype of VADRM application model for typical video analytics applications (i.e., surveillance cameras) is implemented by Convolutional neural network (CNN). The analytical model is created based on the workload characterization of the prototype and used in the general simulation to measure the effectiveness of VADRM for employing edge computing instead of the cloud. The results show VADRM can allocate 45% of the data size for edge processing and 55.50% for cloud processing. iFogSim toolkit is used to simulate the fog environment and measure network performance when using VADRM model.

关键词： Video Analytics Fog Computing Edge Computing iFogSim Convolutional neural network

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：