检索结果-内蒙古大学图书馆

作者： Zhu, Suguo Fang, Zhenying Gao, Fei Key Laboratory of Complex Systems Modeling and Simulation School of Computer Science and Technology Hangzhou Dianzi University Hangzhou310018 China State Key Laboratory of Integrated Services Networks Xidian University Xian710071 China

Recently, deep learning is widely developed in computer vision applications. In this paper, a novel simple tracker with deep learning is proposed to complete the tracking task. A simple fully convolutional Siamese network is applied to capture the similarity between different frames. Nevertheless, the detailed information from lower layers, which is also important for locating the target object, is not considered into the tracking task. In this paper, the detailed information from two lower layers is considered into the response map to improve the performance and not to increase much time spent. This leads more significant improvement for feature representation and localization of the target object. The experimental results demonstrate that the proposed algorithm is efficient and robust compared with the baseline and the state-of-the-art trackers. © 2018, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep Modular Co-Attention networks for visual question answering

arXiv

引用

arXiv 2019年

作者： Yu, Zhou Yu, Jun Cui, Yuhao Tao, Dacheng Tian, Qi Key Laboratory of Complex Systems Modeling and Simulation School of Computer Science and Technology Hangzhou Dianzi University UBTECH Sydney AI Centre School of Computer Science FEIT University of Sydney Noah's Ark Lab Huawei

Visual Question Answering (VQA) requires a finegrained and simultaneous understanding of both the visual content of images and the textual content of questions. Therefore, designing an effective 'co-attention' model to associate key words in questions with key objects in images is central to VQA performance. So far, most successful attempts at co-attention learning have been achieved by using shallow models, and deep co-attention models show little improvement over their shallow counterparts. In this paper, we propose a deep Modular Co-Attention Network (MCAN) that consists of Modular Co-Attention (MCA) layers cascaded in depth. Each MCA layer models the self-attention of questions and images, as well as the guided-attention of images jointly using a modular composition of two basic attention units. We quantitatively and qualitatively evaluate MCAN on the benchmark VQA-v2 dataset and conduct extensive ablation studies to explore the reasons behind MCAN's effectiveness. Experimental results demonstrate that MCAN significantly outperforms the previous state-ofthe- A rt. Our best single model delivers 70.63% overall accuracy on the test-dev set. Code is available at http://***/MILVLG/mcan-vqa. Copyright © 2019, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

MobileEdge: Enhancing On-Board Vehicle Computing Units Using Mobile Edges for CAVs

MobileEdge: Enhancing On-Board Vehicle Computing Units Using...

引用

International Conference on Parallel and Distributed systems (ICPADS)

作者： Lin Wang Qingyang Zhang Youhuizi Li Hong Zhong Weisong Shi School of Computer Science and Technology Anhui University Hefei China Key Laboratory of Complex Systems Modeling and Simulation Hangzhou Dianzi University Hangzhou China Department of Computer Science Wayne State University Detroit MI USA

ISBN: (纸本)9781728125848

As the rapid growth of connected and autonomous vehicles (CAVs) and 5G intensifies, more third-party applications are increasingly being deployed on CAVs. They not only improve user experience but also provide more helpful services, for example, enhancing public safety by recognizing criminals in real-time videos. Current CAVs prefer to process collected data on the vehicle to avoid long transmission latency and extra network cost. However, due to the limitations of the on-board vehicle computing unit (VCU) and increasing use of computing-intensive in-vehicle applications, the burden of on-board VCU has sharply increased, which may affect driving safety. In particular, for existing vehicles on the road, adding more computing devices is a challenge if not impossible due to cost concerns. Inspired by edge computing, we propose a novel platform, MobileEdge, to enhance the computing capability of the unchangeable on-board VCU, which leverages mobile devices as edge nodes, e.g., the passengers' smartphones, by offloading computing tasks to them for collaboratively computing. Moreover, MobileEdge provides the dynamic management of mobile devices, monitoring device status and interfaces for customizable task offloading strategies and eventually achieves optimal task scheduling. We build a prototype to demonstrate the designed platform and evaluate three task offloading strategies which were implemented based on the developed interfaces. The results show that MobileEdge significantly reduces the application response latency. Compared with the baseline which does not employ task offloading, the response latency is almost near real-time when more computing resources are available. In addition, the proposed shortest response latency strategy outperforms the best overall task scheduling among the three strategies.

关键词： Edge computing Vehicular data analysis Distributed computing Connected and autonomous vehicles Distributed computing TASK SCHEDULING Response Latency UNLOADING vehicle vinyl chloride-urethane copolymer Mobile devices unchangeable quasi-realtime language Public safety

来源：评论

学校读者我要写书评

暂无评论

Multi-view adaptive fusion network for 3D object detection

arXiv

引用

arXiv 2020年

作者： Wang, Guojun Tian, Bin Zhang, Yachen Chen, Long Cao, Dongpu Wu, Jian State Key Laboratory of Automotive Simulation and Control Jilin University Changchun130022 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100190 China School of Data and Computer Science Sun Yat-Sen University Guangzhou510275 China Lab University of Waterloo Canada College of Automotive Engineering Jilin University China

3D object detection based on LiDAR-camera fusion is becoming an emerging research theme for autonomous driving. However, it has been surprisingly difficult to effectively fuse both modalities without information loss and interference. To solve this issue, we propose a single-stage multi-view fusion framework that takes LiDAR Birds-Eye View, LiDAR Range View and Camera View images as inputs for 3D object detection. To effectively fuse multi-view features, we propose an Attentive Pointwise Fusion (APF) module to estimate the importance of the three sources with attention mechanisms which can achieve adaptive fusion of multi-view features in a pointwise manner. Besides, an Attentive Pointwise Weighting (APW) module is designed to help the network learn structure information and point feature importance with two extra tasks: foreground classification and center regression, and the predicted foreground probability will be used to reweight the point features. We design an end-to-end learnable network named MVAF-Net to integrate these two components. Our evaluations conducted on the KITTI 3D object detection datasets demonstrate that the proposed APF and APW module offer significant performance gain and that the proposed MVAF-Net achieves state-of-the-art performance in the KITTI benchmark. Copyright © 2020, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Power Characterization of Memory Intensive Applications: Analysis and Implications 1st

Power Characterization of Memory Intensive Applications: Ana...

引用

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

作者： Qiu, Yeliang Jiang, Congfeng Fan, Tiantian Wang, Yumei Zhang, Liangbin Wan, Jian Shi, Weisong Key Laboratory of Complex Systems Modeling and Simulation Ministry of Education Hangzhou Dianzi University Hangzhou310037 China School of Computer Science and Technology Hangzhou Dianzi University Hangzhou310037 China College of Big Data and Software Engineering Zhejiang Wanli University Ningbo China School of Information and Electronic Engineering Zhejiang University of Science and Technology Hangzhou310023 China Department of Computer Science Wayne State University DetroitMI48202 United States

ISBN: (纸本)9783030328122

DRAM is a significant source of server power consumption especially when the server runs memory intensive applications. Current power aware scheduling assumes that DRAM is as energy proportional as other components. However, the non-energy proportionality of DRAM significantly affects the power and energy consumption of the whole server system when running memory intensive applications. Thus good knowledge of server power characterization under memory intensive workloads can help better workload placement with power reduction. In this paper, we investigate the power characteristics of memory intensive applications on real rack servers of different generations. Through comprehensive analysis we find that (1) Server power consumption changes with workload intensity and concurrent execution threads. However, fully utilized memory systems are not the most energy efficient. (2) Powered memory modules of installed memory capacity, i.e. the memory capacity per processor core has significant impact on the application’s performance and server power consumption even if the memory system is not fully utilized. (3) Memory utilization is not always a good indicator for server power consumption when it is running memory intensive applications. Our experiments show that hardware configuration, workload types, as well as concurrently running threads have significant impact on a server’s energy efficiency when running memory intensive applications. Our findings presented in this paper provide useful insights and guidance to system designers, as well as data center operators for energy efficiency aware job scheduling and power reductions. © 2019, Springer Nature Switzerland AG.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Scalability Evaluation of Big Data Processing Services in Clouds 1st

Scalability Evaluation of Big Data Processing Services in Cl...

引用

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

作者： Zhou, Xin Jiang, Congfeng Qiu, Yeliang Fan, Tiantian Wang, Yumei Zhang, Liangbin Wan, Jian Shi, Weisong Key Laboratory of Complex Systems Modeling and Simulation Ministry of Education Hangzhou Dianzi University Hangzhou310037 China School of Computer Science and Technology Hangzhou Dianzi University Hangzhou310037 China College of Big Data and Software Engineering Zhejiang Wanli University Ningbo China School of Information and Electronic Engineering Zhejiang University of Science and Technology Hangzhou310023 China Department of Computer Science Wayne State University DetroitMI48202 United States

ISBN: (纸本)9783030328122

Currently, many cloud providers deploy their big data processing systems as cloud services, which helps users conveniently manage and process their data in clouds. Among different service providers’ big data processing services, how to evaluate and compare their scalability is an interesting and challenging work. Most traditional benchmark tools focus on performance evaluation of big data processing systems, such as aggregated throughput and IOPS, but fail to conduct a quantitative analysis of their scalability. In this paper, we propose a measurement methodology to quantify the scalability of big data processing services, which makes the cloud services scalability comparable. We conduct a group of comparative experiments on AliCloud E-MapReduce and Baidu MRS, and collect their respective scalability characteristics under Hadoop and Spark workloads. The scalability characteristics observed in our work could help cloud users choose the best cloud service platform to set up an optimized big data processing system to achieve their specific goals more successfully. © 2019, Springer Nature Switzerland AG.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Multimodal unified attention networks for vision-and-language interactions

arXiv

引用

arXiv 2019年

作者： Yu, Zhou Cui, Yuhao Yu, Jun Tao, Dacheng Tian, Qi Key Laboratory of Complex Systems Modeling and Simulation School of Computer Science and Technology Hangzhou Dianzi University China UBTECH Sydney AI Centre School of Computer Science Faculty of Engineering University of Sydney Australia Noah's Ark Lab Huawei China

Learning an effective attention mechanism for multimodal data is important in many vision-and-language tasks that require a synergic understanding of both the visual and textual contents. Existing state-of-the-art approaches use co-attention models to associate each visual object (e.g., image region) with each textual object (e.g., query word). Despite the success of these co-attention models, they only model inter-modal interactions while neglecting intra-modal interactions. Here we propose a general 'unified attention' model that simultaneously captures the intra- and inter-modal interactions of multimodal features and outputs their corresponding attended representations. By stacking such unified attention blocks in depth, we obtain the deep Multimodal Unified Attention Network (MUAN), which can seamlessly be applied to the visual question answering (VQA) and visual grounding tasks. We evaluate our MUAN models on two VQA datasets and three visual grounding datasets, and the results show that MUAN achieves top level performance on both tasks without bells and whistles. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

DyCREM: Dynamic Credit Risk Management Using Edge-Based Blockchain

DyCREM: Dynamic Credit Risk Management Using Edge-Based Bloc...

引用

IEEE/ACM Symposium on Edge Computing (SEC)

作者： Youhuizi Li Weisong Shi Mohit Kumar Jun Chen Key Laboratory of Complex Systems Modeling and Simulation Hangzhou Dianzi University Hangzhou China Department of Computer Science Wayne State University Detroit USA Chain Oriental LLC Beijing China

ISBN: (纸本)9781538694466;9781538694459

Banks play an important role in the financial market, and their profit mainly comes from loan service. Traditional risk management approaches, as the key part in the loan services, still have several issues including low data credibility, unreliable checking and alert delay. We propose a novel dynamic credit risk management system which leverages edge-based blockchain technology to improve the performance and provide a fair loan environment. Besides, we discuss the potential challenges that the system may face, such as data standards, security on edge and scalability.

关键词： Risk management Edge computing Data models Computational modeling Real-time systems Analytical models

来源：评论

学校读者我要写书评

暂无评论

Energy efficiency comparison of hypervisors 7

Energy efficiency comparison of hypervisors

引用

7th International Green and Sustainable Computing Conference, IGSC 2016

作者： Jiang, Congfeng Ou, Dongyang Wang, Yumei You, Xindong Zhang, Jilin Wan, Jian Luo, Bing Shi, Weisong Key Laboratory of Complex Systems Modeling and Simulation Hangzhou Dianzi University Zhejiang China Mobile and Internet Systems Laboratory Department of Computer Science Wayne State University DetroitMI United States

ISBN: (纸本)9781509051175

Current cloud data centers are fully virtualized for service consolidations and power/energy reduction. Although virtualization could reduce real time power and overall energy consumption, the energy characteristics of hypervisors hosting different workloads are not well profiled and understood. In this paper, we investigate the power and energy characteristics of mainstream hypervisors and container engine, i.e., VMware ESXi, Microsoft Hyper-V, KVM, XenServer and Docker, on five different platforms (two mainstream 2U rack servers, one emerging ARM64 server, one desktop server, and one laptop) with hundreds of hours' power measures. We use both computing intensive and mixed web server-database workloads to explore the power and energy characteristics of different hypervisors. Extensive experiment results of four workload levels (very light, light, fair, and very heavy workload) demonstrate that hypervisors expose different power and energy characteristics. We find that: (1) Hypervisors expose different power and energy consumption on the same hardware running same workloads. (2) Although mainstream hypervisors have different energy efficiencies aligned with different workload types and workload levels, there is no single hypervisor that outperforms all other hypervisors on all platforms in terms of power or energy consumptions. (3) Although container virtualization is considered as light-weight virtualization in terms of implementation and maintenance, it is not essentially more power efficient than traditional virtualization technology. (4) ARM64 server does have lower power consumption, but they finish computing jobs with longer execution time and sometimes consume more energy. And ARM64 servers has medium energy consumption per database operations for mixed workloads. The results presented in this paper provide useful insights to system designers, as well as data center operators for power-aware workload placement and virtual machine scheduling. © 2016 IEEE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Energy efficiency analysis of heterogeneous platforms: Early experiences 7

Energy efficiency analysis of heterogeneous platforms: Early...

引用

7th International Green and Sustainable Computing Conference, IGSC 2016

作者： Li, Youhuizi Shi, Weisong Jiang, Congfeng Zhang, Jilin Wan, Jian Key Laboratory of Complex Systems Modeling and Simulation Hangzhou Dianzi University China Mobile and Internet Systems Laboratory Wayne State University United States School of Information and Electronic Engineering Zhejiang University of Science and Technology China

ISBN: (纸本)9781509051175

Heterogeneous multi-core platforms, e.g., ARM's ***, are a promising trend to improve the performance and energy efficiency of future mobile systems. However, the immediate benefits and the challenges to take advantage of the heterogeneity are still not clear. In this paper, we present our early experiences about the energy efficiency of the two *** heterogeneous platforms: ODROID XU+E and ODROID XU3. We quantified compared them with homogeneous platforms through multiple benchmarks which include popular mobile applications and high-performance parallel benchmarks. Besides, we analyzed the scheduling impact on the energy consumption of the heterogeneous platforms and the migration cost is also discussed. Based on the results, several insights, such as fine-granularity power control and thread level parallelism, related to hardware, application and system design are derived. © 2016 IEEE.

关键词： Energy utilization

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：