Title:
SEMICONDUCTOR NEURAL CIRCUIT DEVICE
Document Type and Number:
WIPO Patent Application WO/1994/010648
Kind Code:
A1
Abstract:
A semiconductor neural circuit device having a very simple circuit and a self-teaching function, by which a neural network is allowed to learn. The device comprises synapse circuits which output weighted values, and neuron circuits which execute linear addition of the output signals from the synapse circuits, and output the signal voltages of high and low levels with respect to a given threshold value VTH. In the case of learning of increasing the total value Z, only when VTH - < Z < VTH + alpha with respect to two positive parameters and alpha, the weighted value of predetermined synapse circuits which input signals to the neuron circuit is increased by a given positive value. Conversely, in the case of learning of decreasing the total value Z, only when the VTH - alpha < Z < VTH + , the weighted value of predetermined synapse circuits which input signals to the neuron circuit is decreased by a given positive value.__________________
Inventors:
SHIBATA TADASHI (JP)
OHMI TADAHIRO (JP)
OHMI TADAHIRO (JP)
Application Number:
PCT/JP1993/001573
Publication Date:
May 11, 1994
Filing Date:
October 29, 1993
Export Citation:
Assignee:
SHIBATA TADASHI (JP)
OHMI TADAHIRO (JP)
OHMI TADAHIRO (JP)
International Classes:
G06G7/60; G06F15/18; G06N3/063; G06N99/00; H01L21/822; H01L27/04; H01L29/78; (IPC1-7): G06G7/60
Foreign References:
| JPH0430280A | 1992-02-03 |
Other References:
See also references of EP 0667588A4
Download PDF: