An associative memory model with a forgetting process a la Mezard et al. is investigated for a piecewise nonmonotonic output function by the SCSNA proposed by Shiino and Fukai. Similar to the formal monotonic two-state model analyzed by Mezard et al. , the discussed nonmonotonic model is also free from a catastrophic deterioration of memory due to overloading. We theoretically obtain a relationship between the storage capacity and the forgetting rate, and find that there is an optimal value of forgetting rate, at which the storage capacity is maximized for the given nonmonotonicity. The maximal storage capacity and capacity ratio (a ratio of the storage capacity for the conventional correlation learning rule to the maximal storage capacity) increase with nonmonotonicity, whereas the optimal forgetting rate decreases with nonmonotonicity.
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
Kazushi MIMURA, Masato OKADA, Koji KURATA, "Associative Memory Model with Forgetting Process Using Nonmonotonic Neurons" in IEICE TRANSACTIONS on Information,
vol. E81-D, no. 11, pp. 1298-1304, November 1998, doi: .
Abstract: An associative memory model with a forgetting process a la Mezard et al. is investigated for a piecewise nonmonotonic output function by the SCSNA proposed by Shiino and Fukai. Similar to the formal monotonic two-state model analyzed by Mezard et al. , the discussed nonmonotonic model is also free from a catastrophic deterioration of memory due to overloading. We theoretically obtain a relationship between the storage capacity and the forgetting rate, and find that there is an optimal value of forgetting rate, at which the storage capacity is maximized for the given nonmonotonicity. The maximal storage capacity and capacity ratio (a ratio of the storage capacity for the conventional correlation learning rule to the maximal storage capacity) increase with nonmonotonicity, whereas the optimal forgetting rate decreases with nonmonotonicity.
URL: https://global.ieice.org/en_transactions/information/10.1587/e81-d_11_1298/_p
Copy
@ARTICLE{e81-d_11_1298,
author={Kazushi MIMURA, Masato OKADA, Koji KURATA, },
journal={IEICE TRANSACTIONS on Information},
title={Associative Memory Model with Forgetting Process Using Nonmonotonic Neurons},
year={1998},
volume={E81-D},
number={11},
pages={1298-1304},
abstract={An associative memory model with a forgetting process a la Mezard et al. is investigated for a piecewise nonmonotonic output function by the SCSNA proposed by Shiino and Fukai. Similar to the formal monotonic two-state model analyzed by Mezard et al. , the discussed nonmonotonic model is also free from a catastrophic deterioration of memory due to overloading. We theoretically obtain a relationship between the storage capacity and the forgetting rate, and find that there is an optimal value of forgetting rate, at which the storage capacity is maximized for the given nonmonotonicity. The maximal storage capacity and capacity ratio (a ratio of the storage capacity for the conventional correlation learning rule to the maximal storage capacity) increase with nonmonotonicity, whereas the optimal forgetting rate decreases with nonmonotonicity.},
keywords={},
doi={},
ISSN={},
month={November},}
Copy
TY - JOUR
TI - Associative Memory Model with Forgetting Process Using Nonmonotonic Neurons
T2 - IEICE TRANSACTIONS on Information
SP - 1298
EP - 1304
AU - Kazushi MIMURA
AU - Masato OKADA
AU - Koji KURATA
PY - 1998
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E81-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 1998
AB - An associative memory model with a forgetting process a la Mezard et al. is investigated for a piecewise nonmonotonic output function by the SCSNA proposed by Shiino and Fukai. Similar to the formal monotonic two-state model analyzed by Mezard et al. , the discussed nonmonotonic model is also free from a catastrophic deterioration of memory due to overloading. We theoretically obtain a relationship between the storage capacity and the forgetting rate, and find that there is an optimal value of forgetting rate, at which the storage capacity is maximized for the given nonmonotonicity. The maximal storage capacity and capacity ratio (a ratio of the storage capacity for the conventional correlation learning rule to the maximal storage capacity) increase with nonmonotonicity, whereas the optimal forgetting rate decreases with nonmonotonicity.
ER -