检索结果-内蒙古大学图书馆

Simulation trained CNN for accurate embedded crack length, location, and orientation from ultrasound measurements

INTERNATIONAL JOURNAL OF SOLIDS AND STRUCTURES 2022年 242卷 111521-111521页

作者： Niu, Sijun Srivastava, Vikas Brown Univ Sch Engn Providence RI 02912 USA Brown Univ Ctr Biomed Engn Providence RI 02912 USA

Accurate quantitative characterization of crack length, location, and orientation are critical for the safety assessment of load bearing structures to avoid catastrophic structural failures. Ultrasound non-destructive evaluation is one of the key methods to detect and evaluate embedded flaws inside a material during fabrication or operation. Although significant progress has been made in developing advanced ultrasound sensors and signal data processing methods, current practices rely on human expertise to evaluate the ultrasound measurements, which leads to high uncertainty and errors in the predictions. Here we demonstrate that an embedded crack reflected ultrasound time signal contains complete information about the key characteristics of the crack, which can be accurately quantified using an optimally trained machine learning model. A lack of sufficiently large, well distributed, and suitably labeled datasets to train machine learning models continues to be a significant obstacle for evaluating non-visible cracks. To overcome this limitation, we demonstrate that our finite element simulation trained convolutional neural network (CNN) is able to accurately predict all three crack characteristics from experimentally measured ultrasound non-destructive test signals. We created a moderate size A-scan time signal simulation dataset (1200 scans) for three-dimensional (3D) elliptical penny shaped cracks inside rectangular cuboid steel to train our CNN. Independent validation experiments were performed by conducting 21 ultrasound tests on 3D printed steel specimens containing a variety of embedded crack geometries. We show that our purely finite element simulation trained CNN accurately predicts crack length, crack location, and crack orientation from experimentally measured signals with an average error of 5.7%, 5.6%, and 8.4% for length, location, and orientation, respectively. This approach of utilizing simulation based training of a neural network can be used in othe

关键词： Crack Embedded flaw Non-destructive test NDE Ultrasound Finite element simulation Convolutional neural network CNN Machine learning

来源：评论

学校读者我要写书评

暂无评论

EDGES: An Efficient distributed Graph Embedding System on GPU Clusters

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2021年第7期32卷 1892-1902页

作者： Yang, Dongxu Liu, Junhong Lai, Junjie NVIDIA Beijing 100102 Peoples R China

Graph embedding training models access parameters sparsely in a "one-hot" manner. Currently, the distributed graph embedding neural network is learned by data parallel with the parameter server, which suffers significant performance and scalability problems. In this article, we analyze the problems and characteristics of training this kind of models on distributed GPU clusters for the first time, and find that fixed model parameters scattered among different machine nodes are a major limiting factor for efficiency. Based on our observation, we develop an efficient distributed graph embedding system called EDGES, which can utilize GPU clusters to train large graph models with billions of nodes and trillions of edges using data and model parallelism. Within the system, we propose a novel dynamic partition architecture for training these models, achieving at least one half of communication reduction compared to existing training systems. According to our evaluations on real-world networks, our system delivers a competitive accuracy for the trained embeddings, and significantly accelerates the training process of the graph node embedding neural network, achieving a speedup of 7.23x and 18.6x over the existing fastest training system on single node and multi-node, respectively. As for the scalability, our experiments show that EDGES obtains a nearly linear speedup.

关键词： Training Data models Servers Graphics processing units Computational modeling neural networks Training data Large-scale distributed training graph node embedding GPU clusters parallel algorithm scalability

来源：评论

学校读者我要写书评

暂无评论

G-SLIDE: A GPU-Based Sub-Linear Deep Learning Engine via LSH Sparsification

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2022年第11期33卷 3015-3027页

作者： Pan, Zaifeng Zhang, Feng Li, Hourun Zhang, Chenyang Du, Xiaoyong Deng, Dong Key Lab Data Engn & Knowledge Engn MOE Beijing 100872 Peoples R China Renmin Univ China Sch Informat Beijing 100872 Peoples R China Rutgers State Univ Dept Comp Sci New Brunswick NJ 08901 USA

Deep learning has been one of the trendiest research topics. However, as data quantities rise exponentially, training large neural networks can become prohibitively expensive with billions of parameters. Fortunately, recent research has discovered that not all of the computations in traditional network training are necessary. By selectively sparsifying the majority of the neurons during training, we can still obtain acceptable accuracy. SLIDE, a C++ OpenMP-based sub-linear deep learning engine, has been developed in this situation. SLIDE uses the algorithm of locality sensitive hashing (LSH) to query neurons with high activation in sub-linear time. It achieves a remarkable speedup in training large fully-connected networks by making use of the network sparsity as well as multi-core parallelism. However, SLIDE is limited to CPUs, ignoring the popular GPU devices with greater parallel potential and computational capability. In this article, we propose G-SLIDE, a GPU-based sub-linear deep learning engine, which combines the benefits of SLIDE's adaptive sparsification algorithms with GPUs' high performance. The main challenges in developing G-SLIDE are efficiently using LSH to sparsify networks and training the special sparse neural networks on the GPU. To address these challenges, we propose several novel solutions, such as specific data formats and appropriate workload partitioning for threads to fully utilize the GPU resources. We evaluate G-SLIDE on two extremely sparse datasets with a 2080 Ti GPU, and the results demonstrate that for the time of one training epoch, G-SLIDE can achieve more than 16.4x speedup over SLIDE on a 32-core/64-thread CPU. Furthermore, on the same platform, G-SLIDE can earn an average of 16.2x speedup over TensorFlow-GPU and 30.8x speedup over TensorFlow-CPU.

关键词： Graphics processing units Training Deep learning Neurons Biological neural networks Engines Message systems GPU machine learning system adaptive sparsity sparse neural network LSH

来源：评论

学校读者我要写书评

暂无评论

Higher Reconstruction Performance with Dual Deep neural networks for Wireless Sensor networks Compressed Temporal Signals 12

Higher Reconstruction Performance with Dual Deep Neural Netw...

引用

12th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, IDAACS 2023

作者： Rousseau, Sebastien Negrier, Romain Courreges, Fabien Engoti, Frank Itoua Quere, Raymond University of Limoges Xlim Institute Brive-la-Gaillarde19100 France NAIoBEE Company Brive-la-Gaillarde19100 France

ISBN: (纸本)9798350358056

In many applications using wireless sensor networks, the reliability of monitored data is crucial to analyze situations and take decisions. Compressed sensing methods are effective to ensure durability of a wireless sensor network installation and to overcome communication failures by drastically lowering the amount of data to transmit. The method proposed in this paper aims to perform an abstraction of the signals sparsity model required by compressed sensing in order to use a deep learning-based approach. This method is designed to be effective for all kinds of transient or temporally dynamic signals sufficiently regular to be predictable, that is to say signals that can be implicitly modelled by a sparse representation. Through the collaboration of two recurrent neural networks and signal processing algorithms, our temporal method outperforms, in terms of accuracy, state-of-art methods, including distributed compressed sensing, time series imputation and forecasting deep learning methods. Thus, high compression rates up to 95% can be reached while leading to an error of less than 0.5°C in reconstructing temperature signals. © 2023 IEEE.

关键词： compressed sensing deep learning recurrent neural networks signal processing algorithms wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

AFCNNM: Accelerating Fully Connected neural network on Mesh-based Optical network-on-Chip

AFCNNM: Accelerating Fully Connected Neural Network on Mesh-...

引用

International Symposium on Parallel and distributed processing with Applications, ISPA

作者： Wen Yang Yunting Liao Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang China College of Computer and Information Technology China Three Gorges University Yichang China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

Fully Connected neural network (FCNN) are widely used in image recognition and natural language processing. However, the time cost of training large datasets is high. Optical network-on Chip (ONoC) has been proposed to accelerate the parallel computing of FCNN because of its advantages. Therefore, this paper proposes an accelerated FCNN model based on ONoC. We first design an FCNN-aware mapping strategy, and then propose a group-based inter-core communication scheme with low wavelength requirements according to the distribution of mapping cores. The optimal number of cores in each period is obtained by achieving the trade-off between the communication and computation time. The simulation results show that the proposed scheme has the advantages of low wavelength requirement, short training time and good scalability.

关键词： Training Simulation Computational modeling Scalability neural networks Optical computing network-on-chip Optical fiber networks Parallel processing Optical imaging

来源：评论

学校读者我要写书评

暂无评论

EEG-GCN: Spatio-Temporal and Self-Adaptive Graph Convolutional networks for Single and Multi-View EEG-Based Emotion Recognition

引用

IEEE SIGNAL processing LETTERS 2022年 29卷 1574-1578页

作者： Gao, Yue Fu, Xiangling Ouyang, Tianxiong Wang, Yi Beijing Univ Posts & Telecommun Natl Pilot Software Engn Sch Sch Comp Sci Beijing 100876 Peoples R China Minist Educ Key Lab Trustworthy Distributed Comp & Serv BUPT Beijing 100876 Peoples R China

Graph networks are naturally suitable for modeling multi-channel features of EEG signals. However, the existing study that attempts to utilize graph-based neural networks for EEG-based emotion recognition doesn't take the spatio-temporal redundancy of EEG features and differences in brain topology into account. In this paper, we propose EEG-GCN, a paradigm that adopts spatio-temporal and self-adaptive graph convolutional networks for single and multi-view EEG-based emotion recognition. With spatio-temporal attention mechanism employed, EEG-GCN can adaptively capture significant sequential segments and spatial location information in EEG signals. Meanwhile, a self-adaptive brain network adjacency matrix is designed to quantify the connection strength between the channels, in which way to represent the diverse activation patterns under different emotion scenarios. Additionally, we propose a multi-view EEG-based emotion recognition method, which effectively integrates the diverse features of EEG signals. Extensive experiments conducted on two benchmark datasets SEED and DEAP demonstrate that our proposed method outperforms other representative methods from both single and multiple views.

关键词： Electroencephalography Feature extraction Emotion recognition Convolution Frequency-domain analysis Brain modeling network topology EEG emotion recognition graph convolutional neural network spatio-temporal attention mechanism

来源：评论

学校读者我要写书评

暂无评论

distributed Incentive Algorithm for Fine-Grained Offloading in Vehicular Ad Hoc networks 24th

Distributed Incentive Algorithm for Fine-Grained Offloading...

引用

24th International Conference on Algorithms and Architectures for Parallel processing, ICA3PP 2024

作者： Wu, Junhong Wu, Yalan Huang, Jiale Wu, Jigang School of Computer Science and Technology Guangdong University of Technology Guangzhou China School of Integrated Circuits Guangdong University of Technology Guangzhou China

ISBN: (纸本)9789819615445

In vehicular ad hoc networks, few existing works on task offloading focus on co-offloading at intra-vehicle level and inter-vehicle level for deep neural network (DNN) inference. Moreover, they ignore the decentralized environment and selfishness of vehicles. This seriously limits the improvement of quality of services for DNN inference. To fill this gap, we formulate a fine-grained offloading problem considering two-level offloading and vehicle selfishness. To solve the problem, we propose a distributed algorithm based on coalition game to encourage vehicles to share their resources. Simulation results show that, the proposed algorithm outperforms the state-of-the-arts in terms of system payoff and running time of algorithms for most cases. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Consensus algorithm

来源：评论

学校读者我要写书评

暂无评论

Analysis of the Leaky Integrate-and-Fire neuron model for GPU implementation

引用

JOURNAL OF PARALLEL AND distributed COMPUTING 2022年 163卷 1-19页

作者： Venetis, Ioannis E. Provata, Astero Univ Piraeus Dept Informat Piraeus Greece Natl Ctr Sci Res Demokritos Inst Nanosci & Nanotechnol Athens Greece

Understanding how neurons perform, when they are organized in interacting networks, is a key to understanding how the brain performs complex functions. Different models that approximate the behavior of interconnected neurons have been proposed in the literature. Implementing these models to simulate neuron behavior at an appropriately detailed level to observe collective phenomena is computationally intensive. In this study we analyze the coupled Leaky Integrate-and-Fire model and report on the issues that affect performance when the model is implemented on a GPU. We conclude that the problem is heavily memory-bound. Advances in memory technology at the hardware level seem to be the deciding factor to achieve better performance on the GPU. Our results show that using an NVidia K40 GPU a modest 2x speedup can be achieved compared to a parallel implementation running on a modern multi-core CPU. However, a substantial speedup of 11.1x can be achieved using an NVidia V100 GPU, mainly due to the improvements in its memory subsystem. (C) 2022 Elsevier Inc. All rights reserved.

关键词： Computational neuroscience neural models neural networks Leaky Integrate-and-Fire model GPU processing

来源：评论

学校读者我要写书评

暂无评论

distributed Fixed-Time Algorithms for Time-Varying Constrained Optimization Problems

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年

作者： He, Xing Li, Yue Zhang, Meng Huang, Tingwen Southwest University Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing College of Electronic and Information Engineering Chongqing400715 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

In this paper, the distributed form of the zeroing neural network for solving time-varying optimal problems is put forward. Compared with traditional centralized algorithms, distributed algorithms possess better privacy and scalability. This paper initially proposes a centralized time-varying optimisation algorithm with fixed-time convergence and certain robustness, which is based on the integration-enhanced zeroing neural network. Subsequently, the algorithm is enhanced, and two distributed algorithms are designed separately. Both of these two algorithms have a fixed convergence time and certain robustness. Additionally, this paper utilizes the penalty function approach to handle time-varying optimization problems with inequality constraints, thereby making the algorithm more widely applicable. The effectiveness of the algorithm is verified through several numerical examples, and the applicability of the algorithm is demonstrated by solving the package-level state-of-charge balancing problem. © 2020 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

distributed cooperative neural control of a class of nonlinear multi-agent systems with unknown time-varying control coefficient

引用

INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL processing 2022年第1期36卷 2-16页

作者： Li, Xiaojie Wang, Yiguang Guilin Univ Technol Coll Mech & Control Engn Guilin 541004 Peoples R China

This article studies the leader-follower cooperative tracking problem of a class of multi-agent systems with unknown nonlinear dynamics. As the load of the following agent may be changing throughout the whole work process, we consider the control coefficient of the following agent to be time-varying and nonlinear instead of constant, which is more practical. All agents are connected by the directed communication graph with weighted topology. The followers can have unknown nonidentical nonlinear dynamics and external disturbances. The nonautonomous leader generates the reference trajectory for only part of the followers and others can only receive the information from their neighbors. To achieve the ultimate synchronization of all following agents to the leader, the novel cooperative adaptive control protocols are designed based on the neural approximation and adaptive updating mechanism. A novel singularity-avoided adaptive updating law is proposed to estimate the control coefficient and compensate for the unknown dynamics online. Lyapunov theory is used to prove the ultimate boundedness of the synchronization tracking error. The correctness and effectiveness of the presented control scheme are demonstrated by two simulations in SISO and MIMO cases, respectively.

关键词： cooperative control leader-follower tracking neural network adaptive estimation nonlinear dynamics time-varying control coefficient

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：