This paper presents some experimental results on the realization of a parallel simulation of an Artificial Neural Network (ANN) on a Transputer based system. The Back-Propagation algorithm has been used, and the parallelization technique utilized is the Processor Farm based on the partition of the training patterns between the available transputers so that each transputer (trainee) is trained on a different part of the training set, while another transputer (the controller) keeps track of the global network state
International Conference on Artificial Neural Networks, Amsterdam, 1993
Petridis, V., Adamidis, P. and Margaritis, K. G. (1993). On the Realization of Back-Propagation on a Transputer Based System. In: International Conference on Artificial Neural Networks: conference proceedings, Amsterdam, 1993. Amsterdam: Springer.
Δημοσιεύσεις σε Περιοδικά
Το τεκμήριο πιθανώς υπόκειται σε σχετική με τα Πνευματικά Δικαιώματα νομοθεσία
This item is probably protected by Copyright Legislation
*Institutions are responsible for keeping their URLs functional (digital file, item page in repository site)
Βοηθείστε μας να κάνουμε καλύτερο το OpenArchives.gr.