𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Performance of a backpropagation trained feedforward network on an MIMD architecture

✍ Scribed by Abbas, Hazem M.; Bayoumi, Mohamed M.

Publisher
John Wiley and Sons
Year
1998
Tongue
English
Weight
118 KB
Volume
10
Category
Article
ISSN
1040-3108

No coin nor oath required. For personal study only.

✦ Synopsis

Training of feedforward networks on sequential machines is a computationally expensive process. This has motivated the implementation of parallel versions of the backpropagation training algorithm on different parallel platforms in order to decrease the processing time required for training. In this paper, we are investigating the implementation of backpropagation on the Alex AVX-2 coarse-grained MIMD machine. A master-slave parallel implementation is carried out for the encoder-decoder benchmark problem. A communication model for the broadcasting/gathering is used to study the effect of using different topologies. Then the performance of the backpropagation algorithms is analyzed for different network sizes and numbers of processors when the nodes are arranged as a pipeline array and in a mesh topology.

πŸ“œ SIMILAR VOLUMES

On design of a survivable network archit
On design of a survivable network architecture for dynamic routing: Optimal solution strategy and an efficient heuristic
✍ Iradj Ouveysi; Andrew Wirth πŸ“‚ Article πŸ“… 1999 πŸ› Elsevier Science 🌐 English βš– 262 KB

We investigate network planning and design under volatile conditions of link failures and trac overload. Our model is a non-simultaneous multi-commodity problem, with any particular two link failure being considered as one scenario. We show that the optimal solution model is not practically solvable

Performance of an artificial neural netw
Performance of an artificial neural network on forecasting the daily occurrence and annual depth of rainfall at a tropical site
✍ Akila D. Kumarasiri; Upul J. Sonnadara πŸ“‚ Article πŸ“… 2008 πŸ› John Wiley and Sons 🌐 English βš– 244 KB πŸ‘ 1 views

## Abstract Performance of a feed‐forward back‐propagation artificial neural network on forecasting the daily occurrence and annual depth of rainfall at a single meteorological station is presented. Both short‐term and long‐term forecasting was attempted, with ground level data collected by the met

A simple and efficient approach to train
A simple and efficient approach to train artificial neural networks using a genetic algorithm to calculate the resonant frequency of an RMA on thick substrate
✍ Bonomali Khuntia; Shyam S. Pattnaik; Dhruba C. Panda; Dipak K. Neog; S. Devi; Ma πŸ“‚ Article πŸ“… 2004 πŸ› John Wiley and Sons 🌐 English βš– 249 KB πŸ‘ 1 views

## Abstract Both genetic algorithms (GAs) and artificial neural networks (ANNs) have been used in the field of computational electromagnetics as the most powerful optimizing tools. In this paper, a simple and efficient method is presented to handle the problem of competing convention while training

PERFORMANCE ANALYSIS OF HETEROGENEOUS TR
PERFORMANCE ANALYSIS OF HETEROGENEOUS TRAFFIC ON AN INTEGRATED NETWORK LINK WITH FINITE WAITING ROOM AND ANTICIPATED-RELEASE RESERVATION POLICY
✍ C. S. REECE; W.-J. HSIN; D. MEDHI; A. VAN DE LIEFVOORT πŸ“‚ Article πŸ“… 1996 πŸ› John Wiley and Sons 🌐 English βš– 923 KB

We consider a system where the superposition of two heterogeneous Poisson traffic streams is offered to an integrated network link in synchronous transfer mode, where one stream follows the blockedand-cleared mode ('loss' mode) and the other can wait (finitely) if bandwidth is not available for conn