Title :
GSN neural network in hardware
Author :
Uebel, Luís Felipe ; Barone, Dante
Author_Institution :
Ciencia da Comput., Univ. Federal do Rio Grande do Sul, Porto Alegre, Brazil
Abstract :
Describes a VLSI implementation of the GSN (goal seeking neuron) Boolean neural network model in an integrated circuit. The DIANNE (Digital Implementation of Artificial Neural NEtwork) is fully digital and asynchronous, computing 600 million synapses per second. The circuit has a linear growth of performance with the increase of the number of chips in the neural network. The relationship between area, performance, and number of neurons implemented on the chip presents a good solution to the implementation of artificial neural networks on silicon
Keywords :
VLSI; neural chips; performance evaluation; Boolean neural network model; DIANNE; GSN neural network; VLSI implementation; digital asynchronous neural chip; goal seeking neuron; integrated circuit; performance; Artificial neural networks; Computer networks; Costs; Fabrication; Integrated circuit modeling; Intelligent networks; Neural network hardware; Neural networks; Neurons; Silicon;
Conference_Titel :
Artificial Neural Networks and Expert Systems, 1993. Proceedings., First New Zealand International Two-Stream Conference on
Conference_Location :
Dunedin
Print_ISBN :
0-8186-4260-2
DOI :
10.1109/ANNES.1993.323065
