Associative massively parallel computers
Article Abstract:
Massively Parallel Computers (MPCs) are being used extensively to supply powerful computing environments for a number of computationally intensive applications that need up to Tera (1E12) operations per second (TOPS) performance levels. Associative Processing (AP) architectures are being used in second-generation MPCs because APs provide considerable flexibility for applications and maintain high efficiency. Associative String Processor (ASP), a form of AP, is particularly suitable for second-generation MPCs because it was developed as a Silicon-efficient parallel architecture.
Publication Name: Proceedings of the IEEE
Subject: Electronics
ISSN: 0018-9219
Year: 1991
User Contributions:
Comment about this article or add new information about this topic:
Backpropagation through time: what it does and how to do it
Article Abstract:
Backpropagation is a method for calculating a single quantity's derivatives with respect to a set of input values. Backpropagation through time finds derivatives for dynamic systems. Basic backpropagation is used to teach neural networks by adjusting weights to produce smaller output errors. Backpropagation through time allows networks to account for results from previous determinations. Backpropagation through time can be used in areas such as pattern recognition and systems identification.
Publication Name: Proceedings of the IEEE
Subject: Electronics
ISSN: 0018-9219
Year: 1990
User Contributions:
Comment about this article or add new information about this topic: