Neural Computing and the GALATEA Project
This paper reviews the fundamentals of neural computing which includes: neural network models, neural network programming environments, and neurocomputer, specialised hardware for neural networks.
It then describes the ESPRIT II GALATEA project, and its predecessor PYGMALION, which provide the focus of neural computing research in the European Community. PYGMALION has developed a general programming environment for neural networks, and the goal of GALATEA is to build upon this environment, to produce a comprehensive neurocomputing system. This neurocomputing system will comprise: a sophisticated programming environment capable of mapping a network on to a range of conventional computers, including parallel machines; a novel general-purpose neurocomputer, and an integral silicon compiler for translating a network into VLSI chips.
KeywordsNeural Network Neural Network Model Parallel Machine Neural Computing Artificial Neuron
Unable to display preview. Download preview PDF.
- 1.I. Aleksander and H. Morton, “Introduction to Neural Computing,” in North Oxford Press, 1990.Google Scholar
- 2.G. Shepherd, “Synaptic Organisation of the Brain,” John Wiley & Sons.Google Scholar
- 3.P.D. Wasserman, “Neural Computing: theory & practice,” in Van Nostrand Reinhold, 1989.Google Scholar
- 4.Nestor, “Nestor Development System User’s Guide,” Nestor Inc., 1988.Google Scholar
- 5.Olmsted & Watkins, “Neural Network Library,” Olmsted & Watkins Inc, 1988.Google Scholar
- 8.Hecht Nielsen, “Hecht-Nielsen Neurocomputing ANZA and AXON,” Hecht-Nielsen Neurocomputing, 1988.Google Scholar
- 9.SAIC, “DELTA/SIGMA/ANSim, editorial,” Neurocomputers, vol. 2, no. 1, 1988.Google Scholar
- 10.NEC, “Neuro-07 (in Japanese),” Nikkan Kogyo, February 1989.Google Scholar
- 12.H.P. Graf, L.D. Jackel, and W. Hubbard, “VLSI Implementation of a Neural Network Model,” IEEE COMPUTER, pp. 41–49, March 1988.Google Scholar