检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Chen, Xiaoming Han, Yinhe Wang, Yu Center for Intelligent Computing Systems State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Department of Electronic Engineering Tsinghua University Beijing China

In current convolutional neural network (CNN) accelerators, communication (i.e., memory access) dominates the energy consumption. This work provides comprehensive analysis and methodologies to minimize the communication for CNN accelerators. For the off-chip communication, we derive the theoretical lower bound for any convolutional layer and propose a dataflow to reach the lower bound. This fundamental problem has never been solved by prior studies. The on-chip communication is minimized based on an elaborate workload and storage mapping *** in addition design a communication-optimal CNN accelerator architecture. Evaluations based on the 65nm technology demonstrate that the proposed architecture nearly reaches the theoretical minimum communication in a threelevel memory hierarchy and it is computation dominant. The gap between the energy efficiency of our accelerator and the theoretical best value is only 37-87%. Copyright © 2019, The Authors. All rights reserved.

关键词： Acceleration

来源：评论

学校读者我要写书评

暂无评论

Online Internet Anomaly Detection With High Accuracy: A Fast Tensor Factorization Solution

Online Internet Anomaly Detection With High Accuracy: A Fast...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Xiaocan Li Kun Xie Xin Wang Gaogang Xie Jigang Wen Guangxing Zhang Zheng Qin Department of Electrical and Computer Engineering State University of New York at Stony Brook USA College of Computer Science and Electronics Engineering Hunan University Changsha China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences

Traffic anomaly detection is critical for advanced Internet management. Existing detection algorithms usually work off-line and cannot timely detect anomalies. They also suffer from high cost for storage and computation. Although online and accurate traffic anomaly detection is very important, it very difficult to achieve. We propose to utilize tensor model to well exploit the multi-dimensional information hidden in the traffic data for more accurate online Internet anomaly detection. We decouple the tensor recovery problem to iteratively solve two sub problems, a tensor factorization sub-problem and an anomaly detection sub-problem. To reduce the high cost for computation and storage involved in tensor factorization, we propose two lightweight techniques to effectively derive factor matrices of tensor in the current window and iteration, taking advantage of tensor decomposition results of the previous window and iteration. We have done extensive experiments using two real traffic traces to compare with three tensor based algorithms and three matrix based algorithms. The experiment results demonstrate that our online anomaly detection algorithm can achieve the same anomaly detection accuracy as that of the best offline tensor based algorithm, but at 6100 times faster speed and with very low storage cost.

关键词： Anomaly detection Monitoring Matrix decomposition Microsoft Windows Principal component analysis Data models

来源：评论

学校读者我要写书评

暂无评论

Conceptual model of real-time IoT systems

引用

Frontiers of Information technology & Electronic Engineering 2019年第11期20卷 1457-1464页

作者： Bo YUAN De-ji CHEN Dong-mei XU Ming CHEN Department of Computer Science and Technology Tongji UniversityShanghai 201804China School of Electronics and Information Engineering Jinggangshan UniversityJi'an 343009China MOE Key Laboratory of Embedded System and Service Computing Tongji UniversityShanghai 201804China China Electronics Standardization Institute Beijing 100007China Chinesisch-Deutsche Hochschule for Angewandte Wissenschaften Tongji UniversityShanghai 201804China

We address a special kind of Internet of Things (IoT) systems that are also real-time. We call them real-time IoT (RT-IoT) systems. An RT-IoT system needs to meet timing constraints of system delay, clock synchronization, deadline, and so on. The timing constraints turn to be more stringent as we get closer to the physical things. Based on the reference architecture of IoT (ISO/IEC 30141), the RT-IoT conceptual model is established. The idea of edge subsystem is introduced. The sensing & con-trolling domain is the basis of the edge subsystem, and the edge subsystem usually must meet the hard real-time constraints. The model includes four perspectives, the time view, computation view, communication view, and control view. Each view looks, from a different angle, at how the time parameters impact an RT-IoT system.

关键词： Internet of Things(IoT) Real-time system Conceptual model View Hard/Soft real-time

来源：评论

学校读者我要写书评

暂无评论

A Variable Batch Size Strategy for Large Scale Distributed DNN Training

A Variable Batch Size Strategy for Large Scale Distributed D...

引用

IEEE International Conference on Big Data and Cloud computing (BdCloud)

作者： Zhongzhe Hu Junmin Xiao Zhongbo Tian Xiaoyang Zhang Hongrui Zhu Chengji Yao Ninghui Sun Guangming Tan State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Western Institute of Advanced Technology Chinese Academy of Sciences Chongqing China Engine Megvii Technology Co.Ltd. Beijing China

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

Large batch distributed synchronous stochastic gradient descent (SGD) has been widely used to train deep neural networks on a distributed memory system with multi-nodes, which can leverage parallel resources to reduce the number of iterative steps and speed up the convergence of training process. However, the large-batch SGD leads to a poor test accuracy, which would counteract the benefits of large scale parallel SGD. Existing solutions for large-batch training either significantly degrade accuracy or require massive additional hyper-parameter tuning. To overcome the difficulty above, we propose a novel variable batchsize strategy. With an in-depth analysis of the different stages in the recent multi-step schedule, we find that the training process at the first stage is sensitive to the batchsize, while different batchsize do not significantly impact the latter stages. Based on this discovery, we first claim that different stages of training should use different batchsize. Hence, the variable batchsize strategy is proposed for the large scale distributed training. Furthermore, in order to turn existing hyper-parameters automatically, an auto-tuning engine is designed for the variable batchsize strategy to achieve higher testing accuracy in the extremely large batchsize cases. By using our strategy, we successfully scale the batchsize to 120K in latter stages on ImageNet-1K with ResNet50 without accuracy loss and 128K with slight accuracy loss. The experimental evaluation on 2048 GPUs shows that the variable batchsize strategy with our auto-tuning engine could complete the training of ResNet-50 in 25 minutes. Furthermore, the new strategy successfully decreases the number of parameter updates by about 1.7 times compared with Facebook's multi-step schedule.

关键词： Training Schedules Neural networks Tuning Computational modeling Graphics processing units computer architecture

来源：评论

学校读者我要写书评

暂无评论

Background subtraction on depth videos with convolutional neural networks

arXiv

引用

arXiv 2019年

作者： Wang, Xueying Liu, Lei Li, Guangli Dong, Xiao Zhao, Peng Feng, Xiaobing State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing100049 China College of Computer Science and Technology Jilin University ChangchunJilin130012 China

Background subtraction is a significant component of computer vision systems. It is widely used in video surveillance, object tracking, anomaly detection, etc. A new data source for background subtraction appeared as the emergence of low-cost depth sensors like Microsoft Kinect, Asus Xtion PRO, etc. In this paper, we propose a background subtraction approach on depth videos, which is based on convolutional neural networks (CNNs), called BGSNet-D (BackGround Subtraction neural Networks for Depth video). The method can be used in color unavailable scenarios like poor lighting situations, and can also be applied to combine with existing RGB background subtraction methods. A preprocessing strategy is designed to reduce the influences incurred by noise from depth sensors. The experimental results on the SBM-RGBD dataset show that the proposed method outperforms existing methods on depth data, and even reaches the performance of the methods that use RGB-D data. Copyright © 2019, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Corrigendum to “Trajectory design and resource allocation for UAV energy minimization in a rotary-wing UAV-enabled WPCN” [Alex. Eng. J. 60(1) (2021) 1787–1796]

引用

Alexandria Engineering Journal 2022年第12期61卷 11003-11003页

作者： Zhen Wang Miaowen Wen Shuping Dang Lisu Yu Yuhao Wang School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing 100876 China School of Information Engineering Nanchang University Nanchang 330031 China School of Electronic and Information Engineering South China University of Technology Guangzhou 510641 China Department of Electrical and Electronic Engineering University of Bristol Bristol BS8 1UB UK State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China

来源：评论

学校读者我要写书评

暂无评论

HPC AI500: A Benchmark Suite for HPC AI systems 1st

HPC AI500: A Benchmark Suite for HPC AI Systems

引用

1st International Symposium on Benchmarking, Measuring, and Optimization, Bench 2018

作者： Jiang, Zihan Gao, Wanling Wang, Lei Xiong, Xingwang Zhang, Yuchen Wen, Xu Luo, Chunjie Ye, Hainan Lu, Xiaoyi Zhang, Yunquan Feng, Shengzhong Li, Kenli Xu, Weijia Zhan, Jianfeng State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Dover DE United States Beijing Academy of Frontier Sciences and Technology Beijing China State University of New York Buffalo United States Department of Computer Science and Engineering The Ohio State University Columbus United States National Supercomputing Center in Shenzhen Shenzhen China National Supercomputing Center in Changsha Changsha China National Supercomputing Center in Jinan Jinan China Texas Advanced Computing Center The Texas University at Austin Austin United States

ISBN: (纸本)9783030328122

In recent years, with the trend of applying deep learning (DL) in high performance scientific computing, the unique characteristics of emerging DL workloads in HPC raise great challenges in designing, implementing HPC AI systems. The community needs a new yard stick for evaluating the future HPC systems. In this paper, we propose HPC AI500—a benchmark suite for evaluating HPC systems that running scientific DL workloads. Covering the most representative scientific fields, each workload from HPC AI500 is based on real-world scientific DL applications. Currently, we choose 14 scientific DL benchmarks from perspectives of application scenarios, data sets, and software stack. We propose a set of metrics for comprehensively evaluating the HPC AI systems, considering both accuracy, performance as well as power and cost. We provide a scalable reference implementation of HPC AI500. The specification and source code are publicly available from http://***/HPCAI500/***. Meanwhile, the AI benchmark suites for datacenter, IoT, Edge are also released on the BenchCouncil web site. © 2019, Springer Nature Switzerland AG.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

AIBench: Towards Scalable and Comprehensive Datacenter AI Benchmarking 1st

AIBench: Towards Scalable and Comprehensive Datacenter AI Be...

引用

1st International Symposium on Benchmarking, Measuring, and Optimization, Bench 2018

作者： Gao, Wanling Luo, Chunjie Wang, Lei Xiong, Xingwang Chen, Jianan Hao, Tianshu Jiang, Zihan Fan, Fanda Du, Mengjia Huang, Yunyou Zhang, Fan Wen, Xu Zheng, Chen He, Xiwen Dai, Jiahui Ye, Hainan Cao, Zheng Jia, Zhen Zhan, Kent Tang, Haoning Zheng, Daoyi Xie, Biwei Li, Wei Wang, Xiaoyu Zhan, Jianfeng State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences Beijing China Dover DE United States Beijing Academy of Frontier Sciences and Technology Beijing China University of Chinese Academy of Sciences Beijing China Alibaba Hangzhou China Princeton University Princeton United States Wuba Zhuxi China Tencent Shenzhen China Baidu Beijing China China RISC-V Alliance Beijing China Cambricon Shenzhen China Intellifusion Shenzhen China

ISBN: (纸本)9783030328122

AI benchmarking provides yardsticks for benchmarking, measuring and evaluating innovative AI algorithms, architecture, and systems. Coordinated by BenchCouncil, this paper presents our joint research and engineering efforts with several academic and industrial partners on the datacenter AI benchmarks—AIBench. The benchmarks are publicly available from http://***/AIBench/***. Presently, AIBench covers 16 problem domains, including image classification, image generation, text-to-text translation, image-to-text, image-to-image, speech-to-text, face embedding, 3D face recognition, object detection, video prediction, image compression, recommendation, 3D object reconstruction, text summarization, spatial transformer, and learning to rank, and two end-to-end application AI benchmarks. Meanwhile, the AI benchmark suites for high performance computing (HPC), IoT, Edge are also released on the BenchCouncil web site. This is by far the most comprehensive AI benchmarking research and engineering effort. © 2019, Springer Nature Switzerland AG.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Edge AIBench: Towards comprehensive end-to-end edge computing benchmarking

arXiv

引用

arXiv 2019年

作者： Hao, Tianshu Huang, Yunyou Wen, Xu Gao, Wanling Zhang, Fan Zheng, Chen Wang, Lei Ye, Hainan Hwang, Kai Ren, Zujie Zhan, Jianfeng State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences BenchCouncil [International Open Benchmarking Council Beijing Academy of Frontier Sciences and Technology Chinese University of Hongkong at Shenzhen Zhejiang Lab

In edge computing scenarios, the distribution of data and collaboration of workloads on different layers are serious concerns for performance, privacy, and security issues. So for edge computing benchmarking, we must take an end-to-end view, considering all three layers: client-side devices, edge computing layer, and cloud servers. Unfortunately, the previous work ignores this most important point. This paper presents the BenchCouncil’s coordinated effort on edge AI benchmarks, named Edge AIBench. In total, Edge AIBench models four typical application scenarios: ICU Patient Monitor, Surveillance Camera, Smart Home, and Autonomous Vehicle with the focus on data distribution and workload collaboration on three layers. Edge AIBench is a part of the open-source AIBench project, publicly available from http://***/AIBench/***. We also build an edge computing testbed with a federated learning framework to resolve performance, privacy, and security issues. Copyright © 2019, The Authors. All rights reserved.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

On Theoretical Optimization of the Sensing Matrix for Sparse-Dictionary Signal Recovery

On Theoretical Optimization of the Sensing Matrix for Sparse...

引用

IEEE Global Conference on Signal and Information Processing (GlobalSIP)

作者： Jianchen Zhu Shengjie Zhao Xu Ma Gonzalo R. Arce Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai CHN School of Optics and Photonics Beijing Institute of Technology Beijing CHN Department of Electrical and Computer Engineering University of Delaware Newark Delaware USA

Compressive Sensing (CS) is a new paradigm for the efficient acquisition of signals that have sparse representation in a certain domain. Traditionally, CS has provided numerous methods for signal recovery over an orthonormal basis. However, modern applications have sparked the emergence of related methods for signals not sparse in an orthonormal basis but in some arbitrary, perhaps highly overcomplete, dictionary, particularly due to their potential to generate different kinds of sparse representation of signals. Here, we first propose the Signal space Subspace Pursuit (SSSP) algorithm, and then we derive a low bound on the number of measurements required. The algorithm has low computational complexity and provides high recovery accuracy.

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：