检索结果-内蒙古大学图书馆

Short-term prediction of chaotic time series by using RBF network with regression weights.

International journal of neural systems 2000年第5期10卷 353-364页

作者： Rojas, I. Gonzalez, J. Cañas, A. Diaz, A.F. Rojas, F.J. Rodriguez, M. Department of Architecture and Computer Technology University of Granada Spain

We propose a framework for constructing and training a radial basis function (RBF) neural network. The structure of the gaussian functions is modified using a pseudo-gaussian function (PG) in which two scaling parameters sigma are introduced, which eliminates the symmetry restriction and provides the neurons in the hidden layer with greater flexibility with respect to function approximation. We propose a modified PG-BF (pseudo-gaussian basis function) network in which the regression weights are used to replace the constant weights in the output layer. For this purpose, a sequential learning algorithm is presented to adapt the structure of the network, in which it is possible to create a new hidden unit and also to detect and remove inactive units. A salient feature of the network systems is that the method used for calculating the overall output is the weighted average of the output associated with each receptive field. The superior performance of the proposed PG-BF system over the standard RBF are illustrated using the problem of short-term prediction of chaotic time series.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Extending dynamic reconfiguration to NOWs with adaptive routing 4th

Extending dynamic reconfiguration to NOWs with adaptive rout...

引用

4th International Workshop on Communication, architecture and Applications for Network-based Parallel Computing, CANPC 2000

作者： BermÚdez, Aurelio Alfaro, Francisco J. Casado, Rafael Duato, José Quiles, Francisco J. Sánchez, José L. Department of Computer Engineering and Technology University of Murcia Murcia30071 Spain Department of Computer Science University of Castilla-La Mancha Albacete02071 Spain Department of Information Systems and Computer Architecture Technical University of Valencia Valencia46071 Spain

ISBN: (纸本)3540678794

Many distributed applications executed on networks of workstations (NOWs) require the interconnection network to provide some quality of service (QoS) support. These networks must be able to support topology changes (due to component failures, hot replacement, hot expansion, etc.) without stopping traffic, in order to satisfy QoS requirements. Traditional network reconfiguration methods do not take this into account, causing a serious performance degradation while the network is being reconfigured. In [1,2], we proposed a new dynamic network reconfiguration protocol, called Partial Progressive Reconfiguration. It significantly reduces the negative effects produced by traditional methods. For this reason, it is especially suitable for applications requiring QoS. This reconfiguration protocol requires that messages are routed using up✶/down✶ routing. In this paper, we extend this dynamic reconfiguration technique to support adaptive routing, based on the design methodology for adaptive algorithms proposed in [3,4]. We also present performance evaluation results, clearly showing the benefits of using dynamic reconfiguration combined with adaptive routing. © Springer-Verlag Berlin Heidelberg 2000.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

On the performance of up✶/down✶ routing 4th

On the performance of up✶/down✶ routing

引用

4th International Workshop on Communication, architecture and Applications for Network-based Parallel Computing, CANPC 2000

作者： Alfaro, Francisco J. BermÚdez, Aurelio Casado, Rafael Duato, José Quiles, Francisco J. Sanchez, José L. Department of Computer Engineering and Technology University of Murcia Murcia30071 Spain Department of Computer Science University of Castilla-La Mancha Albacete02071 Spain Department of Information Systems Computer Architecture Technical University of Valencia Valencia46071 Spain

ISBN: (纸本)3540678794

Networks of Workstations (NOWs) are usually arranged as a set of interconnected switches with hosts connected to switch ports through interface cards. Several commercial interconnects for high-speed NOWs use up✶/down✶ routing. Every time the network is powered on or the topology is changed, a configuration algorithm is executed, which provides information about the topology and generates a directed graph. Routing tables are computed from this directed graph. There are several ways to obtain the directed graph. The most frequent way is by means of algorithms based on minimum-depth spanning-trees (MDST) or propagation-order spanning-trees (POST). This paper shows that, for most networks, graphs obtained by means of these methods can be improved in order to achieve higher network performance. © Springer-Verlag Berlin Heidelberg 2000.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

Simulation of hardware support for OpenGL graphics architecture 1

Simulation of hardware support for OpenGL graphics architect...

引用

1st International Conference on Information technology: Coding and Computing, ITCC 2000

作者： Paltashev, T. Govind, N. Abla, G. Department of Electrical and Computer Engineering SD School of Mines and Technology United States Intel Corporation Computing Enhancement Architecture Lab. United States

ISBN: (纸本)0769505406

The growing number of applications for 3D graphics and imaging systems in the mass market requires the customized approach to the design of high-performance 3D graphics and imaging system architectures. That fact coupled with the strong trends to industrial standardization of graphics API, such as OpenGL, leads to enhanced portability for imaging and graphics applications. The progress on the hardware support of API functionality gives a real possibility to use a top-down approach in the design of custom graphics/imaging systems based on the mixed hardware/software implementation of the OpenGL architecture. Software simulation is a relatively cheap and fast way for initial graphics architecture template development and evaluation to acquire the structure and parameters that can be used later for HDL elaboration and simulation. This paper discusses a 3D graphics and imaging system architecture model implementation based on OpenGL API hardware support simulation. The object-oriented approach has been used for this development of analytic and software models simulating hardware units on different stages of the graphics pipeline. The Interactive Imaging System architecture Composer has been developed in this project for fast run-time simulation management, data acquisition and interpretation. The expandable set of architecture templates uses a reconfigurable shared library of "virtual hardware units" simulating the different hardware structures for graphics and image processing. © 2000 IEEE.

关键词： Imaging systems

来源：评论

学校读者我要写书评

暂无评论

A new radial basis function networks structure: application to time series prediction

A new radial basis function networks structure: application ...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： I. Rojas H. Pomares J. Gonzalez E. Ros M. Salmeron J. Ortega A. Prieto Department of Architecture and Computer Technology University of Granada Spain

Describes a structure to create a RBF neural network. This structure has 4 main characteristics. The first one is that the special RBF network architecture uses regression weights to replace the constant weights normally used. These regression weights are assumed to be functions of input variables. The second characteristic is the normalization of the activation of the hidden neurons (weighted average) before aggregating the activations, which, as observed by various authors, produces better results than the classical weighted sum architecture. The third aspect is that a new type of nonlinear function is proposed, the pseudo-gaussian function (PGBF). With this, the neural system gains flexibility, as the neurons possess an activation field that does not necessarily have to be symmetric with respect to the centre or to the location of the neuron in the input space. In addition to this new structure, we propose, as the fourth and final feature, a sequential learning algorithm, which is able to adapt the structure of the network, with this, it is possible to create new hidden units and also to detect and remove inactive units.

关键词： Radial basis function networks Neurons Neural networks Application software computer architecture Input variables Function approximation computer networks Technological innovation

来源：评论

学校读者我要写书评

暂无评论

What are the main functional blocks involved in the design of adaptive neuro-fuzzy inference systems?

What are the main functional blocks involved in the design o...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： I. Rojas J.L. Bernier R. Rodriguez-Alvarez Z. Prieto Department of Architecture and Computer Technology University of Granada Spain

Because there are many possibilities for the set of basic functions, parameters and operators used in the design of an adaptive-network-based fuzzy inference system (ANFTS), the search for the most suitable operators and functional blocks, together with their characterization and evaluation, is also an important topic in the field of neuro-fuzzy design. As shown in papers dealing with real applications, the designer has to select the operator to be used in each phase in the design of a neuro-fuzzy system, and this decision is usually taken in terms of the most common operations performed. Nevertheless, it is very important to determine which factors have the greatest influence on the behaviour and performance of the neuro-fuzzy system. Therefore, the designer should pay close attention to the phase in which the selection of the operator is most statistically significant. In this way, it is possible to obviate a detailed analysis of different configurations that lead to systems with very similar performance. In order to perform this analysis, an appropriate statistical tool has been used: the multifactorial analysis of the variance (ANOVA) which consists of a set of statistical techniques that enable the analysis and comparison of experiments, by describing the interactions and interrelations between either the quantitative or the qualitative variables of the neural network system. By applying this methodology to a great variety of neuro-fuzzy systems, it is possible to obtain general results about the most relevant factors defining the neural network design.

关键词： Adaptive systems Fuzzy systems Analysis of variance Fuzzy neural networks Performance analysis Neural networks Network topology Fuzzy sets Programmable control Adaptive control

来源：评论

学校读者我要写书评

暂无评论

Clock rate versus IPC: the end of the road for conventional microarchitectures 00

Clock rate versus IPC: the end of the road for conventional ...

引用

Proceedings of the 27th annual international symposium on computer architecture

作者： Vikas Agarwal M. S. Hrishikesh Stephen W. Keckler Doug Burger Computer Architecture and Technology Laboratory Department of Computer Sciences The University of Texas at Austin

ISBN: (纸本)9781581132328

The doubling of microprocessor performance every three years has been the result of two factors: more transistors per chip and superlinear scali ng of the processor clock with technology generation. Our results show that, due to both diminishing improvements in clock rates and poor wire scaling as semiconductor devices shrink, the achievable performance growth of conventional microarchitectures will slow substantially. In this paper, we describe technology-driven models for wire capacitance, wire delay, and microarchitectural component delay. Using the results of these models, we measure the simulated performance—estimating both clock rate and IPC —of an aggressive out-of-order microarchitecture as it is scaled from a 250nm technology to a 35nm technology. We perform this analysis for three clock scaling targets and two microarchitecture scaling strategies: pipeline scaling and capacity scaling. We find that no scaling strategy permits annual performance improvements of better than 12.5%, which is far worse than the annual 50-60% to which we have grown accustomed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Evolutionary computation visualization: Application to G-PROP 6th

Evolutionary computation visualization: Application to G-PRO...

引用

6th International Conference on Parallel Problem Solving from Nature, PPSN 2000

作者： Romero, G. Arenas, M.G. Castellano, J.G. Castillo, P.A. Carpio, J. Merelo, J.J. Prieto, A. Rivas, V. Department of Architecture and Computer Technology University of Granada Campus de Puentenueva E. Granada18071 Spain Department of Computer Science University of Jaén Avda. Madrid 35 E. Jaén23071 Spain

ISBN: (纸本)9783540410560

Software visualization is an area of computer science devoted to supporting the understanding and effective use of algorithms. The application of software visualization to Evolutionary Computation has been receiving increasing attention during the last few years. In this paper we apply visualization technique to an evolutionary algorithm for multilayer perceptron training. Our goal is to better understand its internal behavior in order to improve the evolutionary part of the method. As a result of applying this this technique several deficiencies in the method have been discovered. © Springer-Verlag Berlin Heidelberg 2000.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

A network-computing infrastructure for tool experimentation applied to computer architecture education

A network-computing infrastructure for tool experimentation ...

引用

2000 Workshop on computer architecture Education, WCAE 2000

作者： Figueiredo, Renato J. Fortes, José A. B. Eigenmann, Rudolf Kapadia, Nirav Adabala, Sumalatha Miguel-Alonso, Jose Taylor, Valerie Livny, Miron Vidal, Luis Chen, Jan-Jo School of ECE Purdue University United States Department of EECS Northwestern University United States Dept. of Mathematics and Computer Science Chicago State University United States Dept. of Computer Architecture and Technology University of the Basque Country Spain Computer Sciences Department University of Wisconsin-Madison United States

ISBN: (纸本)9781450347303

computer architects increasingly depend on the use of software tools to evaluate and investigate the design of computer systems. It is therefore very important that educators in this field promote extensive tool-based student experimentation in architecture classes. However, the integration of today's complex architecture tools into curricula poses several challenges to an instructor, including management of powerful computing resources, software installation and maintenance, and development of tool-specific educational material. This paper describes how these challenges are addressed by a universally accessible network-computing infrastructure - NETCARE - that provides educators with a Web portal to access computing resources, executable tools and educational material. Copyright held by author.

关键词： Portals

来源：评论

学校读者我要写书评

暂无评论

Clock rate versus IPC: the end of the road for conventional microarchitectures

Clock rate versus IPC: the end of the road for conventional ...

引用

Annual International Symposium on computer architecture, ISCA

作者： V. Agarwal M.S. Hrishikesh S.W. Keckler D. Burger Computer Architecture and Technology Laboratory Department of Computer Sciences University of Texas Austin Austin TX USA

The doubling microprocessor performance every three years has been the result of two factors: more transistors per chip and superlinear scaling of the processor dock with technology generation. Our results show that, due to both diminishing improvements in clock rates and poor wire scaling as semiconductor devices shrink, the achievable performance growth of conventional microarchitectures will slow substantially. In this paper, we describe technology-driven models for wire capacitance wire delay, and microarchitectural component delay. Using the results of these models, we measure the simulated performance-estimating both clock rate and IPC-of an aggressive out-of-order microarchitecture as it is scaled from a 250 nm technology to a 35 nm technology. We perform this analysis for three clock scaling targets and two microarchitecture scaling strategies: pipeline scaling and capacity scaling. We find that no scaling strategy permits annual performance improvements of better than 12.5% which is far worse than the annual 50-60% to which we have grown accustomed.

关键词： Clocks Roads Microarchitecture Wire Delay Microprocessors Transistors Semiconductor devices Capacitance Performance evaluation

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：