Outline of a theory of massively parallel analog computation

International 1989 Joint Conference on Neural Networks Pub Date : 1989-06-01 DOI:10.1109/IJCNN.1989.118390

B. MacLennan

引用次数: 3

Abstract

Summary form only given. The author proposes the following definition of massive parallelism. A computational system is massively parallel if the number of processing elements is so large that it may conveniently be considered a continuous quantity. The author proposes this definition of massive parallelism for a number of reasons. First, skillful behavior seems to require significant neural mass. Second, he is interested in computers, such as optical computers and molecular computers, for which the number of processing elements is effectively continuous. Third, continuous mathematics is generally easier than discrete mathematics. The author develops a theoretical framework for understanding massively parallel analog computers.<>

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

大规模并行模拟计算理论概述

只提供摘要形式。作者提出了大规模并行的以下定义。如果处理元素的数量如此之大，以至于可以方便地认为是连续的数量，则计算系统是大规模并行的。作者提出这个大规模并行定义的原因有很多。首先，熟练的行为似乎需要大量的神经。其次，他对计算机很感兴趣，比如光学计算机和分子计算机，这些计算机的处理元素的数量实际上是连续的。第三，连续数学通常比离散数学更容易。作者发展了一个理解大规模并行模拟计算机的理论框架。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

International 1989 Joint Conference on Neural Networks

自引率

0.00%

发文量

期刊最新文献

Hybrid distributed/local connectionist architectures A new back-propagation algorithm with coupled neuron A novel objective function for improved phoneme recognition using time delay neural networks Optimization of a digital neuron design Multitarget tracking with an optical neural net using a quadratic energy function