检索结果-内蒙古大学图书馆

Outline of Remarks made in Presentation on the Nature, Design, and use of Intelligent Software Agents

IFAC Proceedings Volumes 2000年第28期33卷 13-16页

作者： Yoh-Han Pao Case Western Reserve University Cleveland OH 44106 U.S.A.

There is genuine need for computer software systems that can carry out significant cognitive tasks in robust, autonomous and adaptive manner. There is much confusion in the manner in which such activities are discussed. This is unfortunate because exaggeration can lead to loss of insight and of appreciation of opportunities. In this presentation, we delineate some qualities of intelligent software agents and cite some unique characteristics of such entities. We also attempt to give an overall view of the technologies that support such capabilities.

关键词： Software Agents Learning Design Neural Networks parallel distributed processing

来源：评论

学校读者我要写书评

暂无评论

parallel pathways mediating manual dexterity in the macaque

引用

EXPERIMENTAL BRAIN RESEARCH 1999年第1-2期128卷 101-108页

作者： Darian-Smith, I Burman, K Darian-Smith, C Univ Melbourne Dept Anat & Cell Biol Brain Res Lab Parkville Vic 3052 Australia

Transmission of information along appropriately structured parallel pathways ensures that a great deal of information can be transferred from the source to the target very quickly, and with great security-essential features of any motor control system. Studies over the last two decades have established that the corticospinal and corticocerebellar pathways mediating manual dexterity in the primate are structurally organized to sustain the parallel transmission of sensorimotor information in multiple pathways. Serial, hierarchical control systems now seem insufficient to regulate voluntary hand movements. To achieve the required coordination, and precision and speed of execution, they must be combined with parallel control systems, which themselves incorporate elaborate feedforward and feedback controls. To illustrate these issues, two aspects of the structural organization of parallel sensorimotor pathways mediating manual dexterity in the macaque are reviewed. First, we examine the structure of the multiple corticospinal neuron sub-populations projecting from different areas of the frontoparietal cortex and how they are modified following hemisection of the cervical spinal cord. The remarkable recovery of hand function following spinal hemisection, despite the absence of any structural 'bridging' of the interrupted spinal pathways, and the fact that this is accountable in a parallel but not in a purely serial transmission system, are then reviewed. The second aspect of parallel distributed transmission examined is its occurrence within a single population of relay neurons. Our recent structural analysis of the somatic/dendritic organization of rubrospinal neurons in macaque red nucleus is used. The very large dendritic fields of individual neurons, extending over one-third or more of the nucleus, provide a framework for extracting precise somatotopic information from an input population whose axon terminal arbors overlap extensively, and, which, without effective fi

关键词： primate manual dexterity parallel distributed processing corticospinal pathways spinal cord hemisection red nucleus

来源：评论

学校读者我要写书评

暂无评论

Dopamine and the mechanisms of cognition: Part II. D-amphetamine effects in human subjects performing a selective attention task

引用

BIOLOGICAL PSYCHIATRY 1998年第10期43卷 723-729页

作者： Servan-Schreiber, D Carter, CS Bruno, RM Cohen, JD Univ Pittsburgh Dept Psychiat Pittsburgh PA USA Carnegie Mellon Univ Dept Psychol Pittsburgh PA 15213 USA

Background: A neural network computer model described in a companion paper predicted the effects of increased dopamine transmission on selective attention under two different hypotheses. Methods: To evaluate these predictions we conducted an empirical study in human subjects of D-amphetamine effects on performance of the Eriksen response competition task. Ten healthy volunteers were rested before and after placebo or D-amphetamine in a double-blind crossover design. Results: D-amphetamine induced a speeding of reaction time overall and an improvement of accuracy at fast reaction times but only in the task condition requiring selective attention. Conclusions: This pattern of results conforms to the prediction of the model under the hypothesis that D-amphetamine primarily affects dopamine transmission in cognitive rather than motor networks. This suggests that the principles embodied in parallel distributed processing models of task performance may be sufficient to predict and explain specific behavioral effects of some drug actions in the central nervous system. (C) 1998 Society of Biological Psychiatry. .

关键词： computer models neural networks parallel distributed processing dopamine D-amphetamine information processing gain Eriksen task selective attention

来源：评论

学校读者我要写书评

暂无评论

parallel modelling and neural networks: An overview for transportation land use systems

引用

TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES 1997年第5期5卷 259-271页

作者： Rodrigue, JP Univ Montreal Ctr Res Transportat Quebec City PQ H3C 3J7 Canada

We provide in this conceptual paper an overview of a parallel transportation/land use modelling environment. We argue that sequential urban modelling does not well represent complex urban dynamics. Instead, we suggest a parallel distributed processing structure composed of processors and links between processors. Each processor is a set of neurons and weights between neurons forming a neural network. For spatial systems neural networks have two main paradigms which are processes simulation and pattern association. parallel distributed processing offers a new methodology to represent the relational structure between elements of a transportation/land use system and thus helping to model those systems. We also provide a set of advantages, drawbacks and some research directions about the usage of neural networks for spatial analysis and modelling. (C) 1997 Elsevier Science Ltd. All rights reserved.

关键词： transportation land use modelling parallel distributed processing neural networks

来源：评论

学校读者我要写书评

暂无评论

Non-linear optimization on a parallel Intel i860 RISC based architecture

引用

Computational Economics 1997年第3期10卷 279-294页

作者： Ball, J.F. Dorsey, R.E. Johnson, J.D. Mississippi Center for Supercomputing Research USA Department of Economics and Finance University of Mississippi Conner Hall University U.S.A. Department of Management and Marketing USA

A description of a parallel implementation of the Genetic Algorithm for non-linear optimization is presented. The system is implemented with four Intel i860 RISC processors using INMOS transputers for communications. The performance is compared to the Cray Y-MP, Cray J916, an SGI Power Challenge L R8000 75 MHZ workstation, an SGI Challenge L R4400 100 MHZ work-station, an SGI Indy SC R4400 200 MHZ workstation, and both parastation and Paramid parallel I860 machines, each with one to four participating I860s. Comparisons were made for neural network optimization of a limited dependent variable problem and a chaotic time series. A complex rational expectations model was also estimated. The I860 based machines were able to achieve performances comparable to the supercomputers.

关键词： Genetic algorithms Neural networks parallel distributed processing

来源：评论

学校读者我要写书评

暂无评论

A NEW ELEMENT FOR parallel distributed-processing NETWORK - LEARNING DISTANCE PRODUCT ELEMENT

引用

SYSTEMS AND COMPUTERS IN JAPAN 1994年第6期25卷 104-110页

作者： MURAYAMA, S YOKOI, H KYUSHU INST TECHNOL FAC ENGNKITAKYUSHU 804JAPAN

One of the factors which affect the convergence velocity in the learning by a parallel distributed processing network is the input/output characteristic of its elements. This paper proposes anew a learning distance product element to substitute for the conventional learning threshold elements, and performs the computer simulation of a three-layered network with two inputs and one output using the learning distance product elements. As a result, it is confirmed that on average, learning is accelerated compared to the use of the learning threshold elements.

关键词： NEURAL COMPUTER NEURAL NET parallel distributed processing LEARNING

来源：评论

学校读者我要写书评

暂无评论

COMPARATIVE FAULT-TOLERANCE OF parallel distributed-processing NETWORKS

引用

IEEE TRANSACTIONS ON COMPUTERS 1994年第11期43卷 1323-1329页

作者： SEGEE, BE CARTER, MJ UNIV NEW HAMPSHIRE DEPT ELECT & COMP ENGNDURHAMNH 03824

In this brief contribution, we propose a method for evaluating and comparing the fault tolerance of a wide variety of parallel distributed processing networks (more commonly referred to as artificial neural networks). Despite the fact that these computing networks are biologically inspired and share many features of biological neural networks, they are not inherently tolerant of the loss of processing elements. We examine two classes of networks, multilayer perceptrons and Gaussian radial basis function networks, and show that there is a marked difference in their operational fault tolerance. Furthermore, we show that fault tolerance is influenced by the training algorithm used and even the initial state of the network. Using an idea due to Sequin and Clay, we show that training with intermittent, randomly selected faults can dramatically enhance the fault tolerance of radial basis function networks, while it yields only marginal improvement when used with multilayer perceptrons.

关键词： APPROXIMATION QUALITY BACKPROPAGATION FAULT TOLERANCE FUNCTION APPROXIMATION GENERALIZED RADIAL BASIS FUNCTION NETWORKS INTERMITTENT FAULTS MULTILAYER PERCEPTRON NEURALNETWORKS parallel distributed processing ROBUSTNESS TRAINING METHODS

来源：评论

学校读者我要写书评

暂无评论

Retrieval of clinical science information using an interactive activation and competition network

引用

ARTIFICIAL INTELLIGENCE IN MEDICINE 1996年第4期8卷 359-375页

作者： Cheng, KJ MOUNTAINSIDE FAMILY PRACTICE ASSOC 799 BLOOMFIELD AVE VERONA NJ 07044 USA

A method for storing and retrieving tabulated clinical science information is described. This method uses the interactive activation and competition network and belongs to the realm of parallel distributed processing. The advantages of this method are that information is readily retrievable by name and by any content, and that the best matched information is automatically returned when partially incorrect cues are given. Furthermore, it allows a variable degree of association among different units of information. The basic principles of this method are illustrated through a simple example from clinical bacteriology. In addition, an application of this method to the retrieval of information in cardiology is presented.

关键词： information storage and retrieval parallel distributed processing interactive activation and competition network

来源：评论

学校读者我要写书评

暂无评论

Manual dexterity: How does the cerebral cortex contribute?

引用

CLINICAL AND EXPERIMENTAL PHARMACOLOGY AND PHYSIOLOGY 1996年第10-11期23卷 948-956页

作者： DarianSmith, I Galea, MP DarianSmith, C UNIV MELBOURNE HOWARD FLOREY INST EXPT PHYSIOL & MED PARKVILLE VIC 3052 AUSTRALIA

1. Manual dexterity, of great evolutionary significance to the primates, ranges in complexity from the precise opposition of finger and thumb to Brendal playing Mozart. All dexterity depends on a sustained and rapid transfer of sensorimotor information between the cerebral cortex and the cervical spinal cord. 2. Multiple separate corticospinal neuron populations originate from cortical areas four, the supplementary motor area, anterior cingulate, postarcuate, parietal and insular cortex. Each corticospinal neuron population projects in parallel to all spinal segments, and has a distinctive pattern of terminations. 3. Each corticospinal neuron population has a unique thalamic input which can relay particular sensorimotor information from the sense organs, cerebellum and basal ganglia. The overall structural framework of these sensorimotor pathways, with many parallel corticospinal channels, with interconnections in the cerebral cortex and spinal cord to enable crosstalk between the channels, is that needed for parallel distributed processing, which would enable the very rapid transfer of information between the cerebral cortex and spinal cord needed for any sophisticated use of the hand. 4. Hemisection of the cervical spinal cord in the macaque results in an immediate hemiplegia, with subsequent remarkable although incomplete recovery of hand and finger movements. The only direct corticospinal input to the hemicord caudal to the hemisection, even after 3 years, is the similar to 10% of flbres which cross the midline caudal to the lesion: the fibres 'spared' by the hemisection, A matching 'sparing' of somatosensory input from the paresed limb also occurs. No regeneration of the interrupted pathways has been visualized using modern tracer techniques. 5. Cervical hemisection permanently reduces the number of parallel channels which transmit information between cortex acid spinal cord, but does not reduce their cortical origins nor the neuron populations targeted in the

关键词： cerebral cortex corticospinal neurons macaque monkey parallel distributed processing spinal hemisection

来源：评论

学校读者我要写书评

暂无评论

parallel photomask pattern data conversion and verification system 3

Parallel photomask pattern data conversion and verification ...

引用

Conference on Photomask and X-Ray Mask Technology III

作者： Kanai, Y Shimada, T Sekigawa, K Nishi, M Ishihara, Y Mori, J Akutagawa, S Takahashi, K Miyauchi, T SHINKO ELECT IND CO LTD NAGANO 38122JAPAN

ISBN: (纸本)0819421790

This system is designed to convert and to verify different types of LSI layout design data and EB writer data, such as GDS-II, MEBES, and JEOL, with the help of its original internal format. The internal format data can be processed not only by a single workstation sequentially but also by network-linked workstations or a single workstation having two or more processing units concurrently. This parallel processing mechanism enables the system to reduce total processing time when handling large volume of data. The resultant internal format data produced by the conversion processes including logical operations, magnifications, and re-sizing operations can be immediately verified by built-in DRC program that is driven by a series of user commands, which consequently yields highly reliable output EB writer data. The built-in pattern viewer can graphically visualize any combination of the layout design data, the EB writer data, and the results of each conversion and verification in the same display window to provide users with an intuitive and easy method for verification. The features of the system and the parallel processing performance are described in this paper.

关键词： mask data conversion system data verification system parallel distributed processing UNIX(R) GDS-II MEBES JEOL

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：