检索结果-内蒙古大学图书馆

2021 IEEE symposium Series on Computational Intelligence, SSCI 2021 - proceedings 2021年

作者： Crockett, Keeley Mostaghim, Sanaz Manchester Metropolitan University United Kingdom Otto von Guericke University Magdeburg Germany

来源：评论

学校读者我要写书评

暂无评论

Serious 3D game over a cluster computing for situated learning of traffic signals 19

Serious 3D game over a cluster computing for situated learni...

引用

proceedings of the 23rd IEEE/acm International symposium on Distributed Simulation and Real Time Applications

作者： Carlos Proaño César Villacís Víctor Proaño Walter Fuertes Mario Almache Margarita Zambrano Fernando Galárraga Universidad de las Fuerzas Armadas- ESPE Sangolquí Ecuador

ISBN: (纸本)9781728129235

this research presents a serious 3D game over a computer cluster that allows the learning of traffic signals so that the child is a responsible driver and pedestrian. For this purpose, we have applied an agile methodology to be able to produce a driving and pedestrian simulation system as a serious game, which includes a three-layer hierarchical architecture (orientation layer, management layer, and control layer). This architecture can be used to simulate three environments that are: a) Driver's environment; b) Pedestrian environment; c) Vehicle traffic observer environment. All these environments are controlled by intelligent agents that allow the user to handle the autonomous longitudinal control, the relative distance, the speed of the cars, the collisions between vehicles, the pedestrian collision, excess of the speed limit, circulation in the opposite direction, and respect of traffic signals. Our solution to traffic problems has been validated by the execution of several tests in public schools, and the results demonstrate that this simulation application encourages learning in road safety education for children between 9 and 12 years old, using a distributed environment with a computational cluster in the cloud.

关键词： serious 3D game driving simulator pedestrian simulation road safety education artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

DAHA: Accelerating GNN Training with Data and Hardware Aware Execution Planning 50th

DAHA: Accelerating GNN Training with Data and Hardware Aware...

引用

50th International Conference on Very Large Data Bases, VLDB 2024

作者： Li, Zhiyuan Jian, Xun Wang, Yue Shao, Yingxia Chen, Lei The Hong Kong University of Science and Technology Hong Kong Shenzhen Institute of Computing Sciences China Beijing University of Posts and Telecommunications China Data Science and Analytics Thrust The Hong Kong University of Science and Technology Guangzhou Hong Kong

Graph neural networks (GNNs) have been gaining a reputation for effective modeling of graph data. Yet, it is challenging to train GNNs efficiently. Many frameworks have been proposed but most of them suffer from high batch preparation cost and data transfer cost for mini-batch training. In addition, existing works have limitations on the device utilization pattern, which results in fewer opportunities for pipeline parallelism. In this paper, we present DAHA, a GNN training framework with data and hardware aware execution planning to accelerate end-to-end GNN training. We first propose a data and hardware aware cost model that is lightweight and gives accurate estimates on per-operation time cost for arbitrary input and hardware settings. Based on the cost model, we further explore the optimal execution plan for the data and hardware with three optimization strategies with pipeline parallelism: (1) group-based in-turn pipelining of batch preparation neural training to explore more optimization opportunities and prevent batch preparation bottlenecks;(2) data and hardware aware rewriting for intra-batch execution planning to improve computation efficiency and create more opportunities for pipeline parallelism;and (3) inter-batch scheduling to further boost the training efficiency. Extensive experiments demonstrate that DAHA can consistently and significantly accelerate end-to-end GNN training and generalize to different message-passing GNN models. © 2024, VLDB Endowment. All rights reserved.

关键词： Cost benefit analysis

来源：评论

学校读者我要写书评

暂无评论

Image retrieval and indexing: a hierarchical approach in computing the distance between textured images

Image retrieval and indexing: a hierarchical approach in com...

引用

IEEE International Conference on Image Processing

作者： R. Stoica J. Zerubia J.M. Francos Department Elec & Computer Eng Ben-Gurion University of the Negev Beersheba Israel INRIA-2004 Route des Lucioles Sophia-Antipolis France

This paper presents a method for indexing and retrieval of multimedia data. The proposed indexing and retrieval strategy is based on the usage of textural information contained in the data imagery components as the indexing keys. On the basis of a 2-D Wold-like decomposition, the texture field is represented as the sum of purely indeterministic, harmonic, and evanescent fields. A new rigorous distance measure between textures which employs their estimated parametric models, is developed. This distance measure is then applied to retrieve multimedia records that contain images with textured segments which are similar to those in a given image. Evaluation of this distance measure is computationally efficient, and hence highly suitable for data base retrieval applications.

关键词： Image retrieval Indexing Information retrieval Multimedia databases Image segmentation Navigation Libraries Labeling Parametric statistics Technological innovation

来源：评论

学校读者我要写书评

暂无评论

Computational intelligence approach to real-world cooperative vehicle dispatching problem

Computational intelligence approach to real-world cooperativ...

引用

International IEEE Conference on Intelligent Systems, IS

作者： K. Hirota Kewei Chen Fangyan Dong Department of Computational Intelligence and Systems Science Institute of Technology Yokohama Tokyo Japan Computing Technology Driving Laboratory Kawasaki Japan

The vehicle dispatching problem for cooperative deliveries from multiple depots (VDP/CD/MD) is an important and difficult problem in the transport industries. In order to solve the VDP/CD/MD, A hierarchical multiplex structure (HIMS/sup ++/) calculation model is proposed. The HIMS/sup ++/ model take advantage of object-oriented modeling, heuristic method, and fuzzy inference in (atomic, molecular, individual) 3 layers, so it can find a utility decision (vehicles plan) close to expert dispatcher. Furthermore, since the vital input parameters are few and the computational engine is packaged into a software component, the HIMS/sup ++/ model is a convenient tool for the VDP/CD/MD. The HIMS/sup ++/ model and its optimization algorithm are implemented as a software component using object-oriented paradigm, and through meta-programming. The performance of the HIMS/sup ++/ model is evaluated through experiments using 3 days oil delivery data taken from an actual dispatching center in Tokyo area. A total of 27 tank lorries are available for daily cooperative deliveries from 3 depots to about 30-60 destinations with different owners are used to save a running cost. The experimental results and the evaluations by human experts confirm that the HIMS/sup ++/ model are better than the results of experienced dispatchers in six evaluation objectives, and can be applied to the planning support system for the VDP/CD/MD. The HIMS/sup ++/ model will be able to cover similar transportation problems in the real-world.

关键词： Computational intelligence Intelligent vehicles Dispatching Object oriented modeling Health information management Software tools Atomic layer deposition Engines Software packages Packaging

来源：评论

学校读者我要写书评

暂无评论

Theoretical and experimental study of spherical arrays

Theoretical and experimental study of spherical arrays

引用

Mediterranean Electrotechnical Conference (MELECON)

作者： N. Burum S. Rupcic Z. Sipus Polytechnic of Dubrovnik Dubrovnik Croatia Faculty of Electrical Engineering University J.J. Strossmayer of Osijek Osijek Croatia Faculty of Electrical Engineering and Computing University of Zagreb Zagreb Croatia

The spectral-domain approach to the analysis of the spherical antenna arrays is studied in the paper. The main advantage of transforming the original problem in the spectral domain is that the original three-dimensional problem is reduced to a spectrum of one-dimensional problems. Therefore, the Green's functions needed for applying the moment method, as well as the elements of the moment method matrix, can be determined in the spectral domain. The main disadvantage lies in numerical problems appearing when the structures with large radii are analyzed. In order to avoid numerical problems a modified vector-Legendre transformation and normalized Legendre polynomials are defined and applied to the solution procedure. The abilities of the method are demonstrated on the analysis of the microstrip patch antennas and waveguide antennas mounted on the spherical supporting structure.

关键词： Moment methods Antenna arrays Green's function methods Transmission line matrix methods Planar arrays Microstrip antennas Microstrip antenna arrays Gratings Nonhomogeneous media Electromagnetic fields

来源：评论

学校读者我要写书评

暂无评论

computing stimulation voltage in a bipolar electrode configuration to avoid side effects during deep brain stimulation

引用

SENSORS AND ACTUATORS A-PHYSICAL 2015年 233卷 9-14页

作者： Chuang, Wei-Yi Chao, Paul C. -P. Chen, Shin-Yuan Tsai, Sheng-Tzung Young, Kuu-Young Ting, Ta-Wei Natl Chiao Tung Univ Dept Elect Engn Hsinchu Taiwan Buddhist Tzu Chi Gen Hosp Dept Neurosurg Hualien Taiwan Tzu Chi Univ Dept Med Hualien Taiwan Yuanpei Univ Dept Biomed Engn Hsinchu Taiwan

Deep brain stimulation (DBS) in the subthalamic nucleus (STN) has been applied for advanced Parkinson's disease (PD). In clinical practices, monopolar configuration is often utilized based on clinical efficiency. Meanwhile, the volume of tissue activated (VTA) by DBS is estimated for controlling and/or limiting the stimulated region. However, side effects like dyskinesia and tremor were accompanied during stimulation. Most studies had found that bipolar stimulation is an alternative approach for side effects avoidance. The goal of this paper is to develop a quantified relationship of stimulation voltage from monopolar to bipolar configuration to improve neural stimulation. In this paper, an electromagnetic finite element model is first built for a patient-specific physiological brain model, which is established by magnetic resonance imaging (MRI) data. The model is then used for finite element analysis (PEA) to estimate VTA with varied electrode voltage in monopolar and bipolar configurations. With the goals to avoid side effects and achieve symptoms suppression, the stimulation voltage of bipolar configuration is successfully computerized based on the electromagnetic PEA simulations. Experiments are conducted successfully to validate simulation results. (C) 2015 Elsevier B.V. All rights reserved.

关键词： Deep brain stimulation (DBS) Volume of tissue activated (VTA) Side effects Stimulation configuration Magnetic resonance imaging (MRI) Finite element analysis (FEA)

来源：评论

学校读者我要写书评

暂无评论

Designing a high performance and fault tolerant multistage interconnection network with easy dynamic rerouting 04

引用

proceedings of the Second international conference on Parallel and Distributed Processing and Applications

作者： Ching-Wen Chen Phui-Si Gan Chih-Hung Chang Department of Computer Science and Information Engineering Chaoyang University of Technology Wufeng Taichung County Taiwan ROC

ISBN: (纸本)9783540241287

Designing a reliable and high performance multistage interconnection network (MIN) should consider the following issues carefully: (1) fault tolerance guarantee; (2) easy schemes and hardware design of rerouting switches; (3) low rerouting resulting in a low collision ratio. In this paper, we present the High Performance Chained Multistage Interconnection Network (HPCMIN) which has one-fault tolerance, destination tag routing for easy rerouting, one rerouting hop, resulting in a low collision ratio. From our simulation results, the HPCMIN results in a lower collision ratio than other dynamic rerouting networks. The HPCMIN is embedded with the indirect binary n-cube network(the ICube network) which is equivalent to many important MINs. Thus, the design methods used in the HPCMIN can be applied to these MINs so that they have the characteristics of the HPCMIN.

关键词： performance collision parallel computing multistage interconnection network (MIN) fault tolerance destination tag routing

来源：评论

学校读者我要写书评

暂无评论

Analyzing the performance of genetically designed short-term traffic prediction models based on road types and functional classes 2004

引用

proceedings of the 17th international conference on Innovations in applied artificial intelligence

作者： Ming Zhong Satish Sharma Pawan Lingras Faculty of Engineering University of Regina Regina SK Canada Dept. of Mathematics and Computing Science Saint Mary's University Halifax NS Canada

ISBN: (纸本)9783540220077

Previous research indicates that the accuracy of short-term traffic prediction has close relationship with road type, and different models should be used for different types of roads. Road types are characterized by trip purposes and trip lengths. However, research for evaluating short-term model's performance based on functional classes is largely absent. Road's functional class is based on the functionality of individual road in a highway network. In this study, genetically designed locally weighed regression (LWR) and time delay neural network (TDNN) models were used to predict short-term traffic for eight rural roads in Alberta, Canada. These roads belong to different functional classes from four trip pattern groups. The influence of functional classes on the performance of short-term prediction models were studied for each type of roads. The results indicate that not only road trip pattern groups but also functional classes have large influence on the accuracy of short-term traffic prediction.

关键词：

来源：评论

学校读者我要写书评

暂无评论

How to decode student bottlenecks to learning in computer science (abstract only) 14

How to decode student bottlenecks to learning in computer sc...

引用

proceedings of the 45th acm technical symposium on Computer science education

作者： Adrian German Suzanne Menzel Joan Middendorf John F. Duncan Indiana University School of Computing and Informatics Bloomington IN USA IUPUI School of Education Indianapolis/Bloomington IN USA

ISBN: (纸本)9781450326056

After thirty years or more of the scholarship of teaching and learning (SoTL) many students still struggle to learn at the university level, while ever more techniques are being developed to help students learn and measure their success. In practice, however, efforts to reshape classes begin with questions such as: "How can I make use of this new technique?" Or, "How can I increase my students' critical thinking?" Such questions are often too broad to provide a clear focal point for designing efficient strategies. They sometimes draw attention to parts of the course not in great need of reform, and generally focus the process on the means (teaching) rather than the end (student learning). A shift is taking place in higher education so that teaching is approached not from content or from the teaching methods, but from the mental processes that are crucial to functioning in a discipline. Decoding the Disciplines (Middendorf & Pace, 2004) is an evidence-based, learner-centered methodology that can conclusively improve teaching by prioritizing what really matters in making meaning in a discipline and boosting learning outcomes in the classroom. It offers a meta-strategy, providing both a framework to define the task at hand and a means to determine which tools would be most useful in accomplishing it. Starting from the bottlenecks, the places where students get stuck, and connecting them with what experts do so they won't get stuck, the crucial mental processes drive the teaching and learning so students can be successful.

关键词： reflective teaching learner-centered methodologies evidence-based teaching

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：