We reveal that the degree of pattern separation in a two-layer random neural net with inhibitory connection has a favorable property for pattern classification. We also consider the effect of the number of connections between the two layers on the above property.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Toyoshi TORIOKA, "Property of the Degree of Pattern Separation in a Two-Layer Random Neural Net with Inhibitory Connections" in IEICE TRANSACTIONS on transactions,
vol. E63-E, no. 8, pp. 590-591, August 1980, doi: .
Abstract: We reveal that the degree of pattern separation in a two-layer random neural net with inhibitory connection has a favorable property for pattern classification. We also consider the effect of the number of connections between the two layers on the above property.
URL: https://global.ieice.org/en_transactions/transactions/10.1587/e63-e_8_590/_p
Copy
@ARTICLE{e63-e_8_590,
author={Toyoshi TORIOKA, },
journal={IEICE TRANSACTIONS on transactions},
title={Property of the Degree of Pattern Separation in a Two-Layer Random Neural Net with Inhibitory Connections},
year={1980},
volume={E63-E},
number={8},
pages={590-591},
abstract={We reveal that the degree of pattern separation in a two-layer random neural net with inhibitory connection has a favorable property for pattern classification. We also consider the effect of the number of connections between the two layers on the above property.},
keywords={},
doi={},
ISSN={},
month={August},}
Copy
TY - JOUR
TI - Property of the Degree of Pattern Separation in a Two-Layer Random Neural Net with Inhibitory Connections
T2 - IEICE TRANSACTIONS on transactions
SP - 590
EP - 591
AU - Toyoshi TORIOKA
PY - 1980
DO -
JO - IEICE TRANSACTIONS on transactions
SN -
VL - E63-E
IS - 8
JA - IEICE TRANSACTIONS on transactions
Y1 - August 1980
AB - We reveal that the degree of pattern separation in a two-layer random neural net with inhibitory connection has a favorable property for pattern classification. We also consider the effect of the number of connections between the two layers on the above property.
ER -