Eugene J. H. Kerckhoffs

Speeding up backpropagation training on a hypercube computer

Abstract Training algorithms for artificial neural networks (such as backpropagation) tend to be very time-consuming. It is therefore obvious to capitalize on the intrinsic parallelism of these systems in order to speed up the computations. This paper describes our simulation of backpropagation neural networks on an NCube/4+ parallel computer and the speed-up performances achieved.

Architecture and Programmature of the MIMD-Structured Delft Parallel Processor

In this paper a parallel processor with a MIMD-structure, the so-called Delft Parallel Processor (DPP), will be presented. The DPP, which has been developed at Delft University of Technology, is operational since spring 1983. The architecture and the developed software for programming the DPP will be discussed. The last part of this paper will be a reflection on the general thoughts about a hierarchical hardware/software computing system with a distributed MIMD-structure.

An Environment for Ill-Defined Systems Research

Ill-defined systems like in environmental, biomedical and biotechnological studies pose several challenging problems when studied through simulation. The impact they have on simulators are touched. Methodology is stretched as well as computer hard-and software needs come into play. A proposed simulator environment currently under investigation at the University of Ghent is described in detail.

ECMS: Its Position in the European Simulation History

This contribution describes some major aspects of the history of European simulation conferences, from the early 1970s of the previous century on. The emphasis is on how the current “European Council for Modelling and Simulation” (ECMS) with its yearly ECMS-conferences has originated in this framework and continued up to now.