The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
La durabilité décrit la capacité d'un appareil à fonctionner correctement dans des conditions imparfaites. Nous avons récemment proposé une nouvelle structure de réseau neuronal appelée « réseau neuronal abordable » (AfNN), dans laquelle les neurones abordables de la couche cachée sont considérés comme les éléments responsables de la propriété de robustesse observée dans le fonctionnement du cerveau humain. Alors que nous avons montré précédemment que les AfNN peuvent encore généraliser et apprendre, nous montrons ici que ces réseaux sont robustes contre les dommages survenant après la fin du processus d'apprentissage. Les résultats confortent l’idée selon laquelle les AfNN incarnent la caractéristique importante de la durabilité. Dans notre contribution, nous étudions la durabilité de l'AfNN lorsque certains neurones de la couche cachée sont endommagés après le processus d'apprentissage.
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.
Copier
Yoko UWATE, Yoshifumi NISHIO, Ruedi STOOP, "Durability of Affordable Neural Networks against Damaging Neurons" in IEICE TRANSACTIONS on Fundamentals,
vol. E92-A, no. 2, pp. 585-593, February 2009, doi: 10.1587/transfun.E92.A.585.
Abstract: Durability describes the ability of a device to operate properly in imperfect conditions. We have recently proposed a novel neural network structure called an "Affordable Neural Network" (AfNN), in which affordable neurons of the hidden layer are considered as the elements responsible for the robustness property as is observed in human brain function. Whereas earlier we have shown that AfNNs can still generalize and learn, here we show that these networks are robust against damages occurring after the learning process has terminated. The results support the view that AfNNs embody the important feature of durability. In our contribution, we investigate the durability of the AfNN when some of the neurons in the hidden layer are damaged after the learning process.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E92.A.585/_p
Copier
@ARTICLE{e92-a_2_585,
author={Yoko UWATE, Yoshifumi NISHIO, Ruedi STOOP, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Durability of Affordable Neural Networks against Damaging Neurons},
year={2009},
volume={E92-A},
number={2},
pages={585-593},
abstract={Durability describes the ability of a device to operate properly in imperfect conditions. We have recently proposed a novel neural network structure called an "Affordable Neural Network" (AfNN), in which affordable neurons of the hidden layer are considered as the elements responsible for the robustness property as is observed in human brain function. Whereas earlier we have shown that AfNNs can still generalize and learn, here we show that these networks are robust against damages occurring after the learning process has terminated. The results support the view that AfNNs embody the important feature of durability. In our contribution, we investigate the durability of the AfNN when some of the neurons in the hidden layer are damaged after the learning process.},
keywords={},
doi={10.1587/transfun.E92.A.585},
ISSN={1745-1337},
month={February},}
Copier
TY - JOUR
TI - Durability of Affordable Neural Networks against Damaging Neurons
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 585
EP - 593
AU - Yoko UWATE
AU - Yoshifumi NISHIO
AU - Ruedi STOOP
PY - 2009
DO - 10.1587/transfun.E92.A.585
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E92-A
IS - 2
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - February 2009
AB - Durability describes the ability of a device to operate properly in imperfect conditions. We have recently proposed a novel neural network structure called an "Affordable Neural Network" (AfNN), in which affordable neurons of the hidden layer are considered as the elements responsible for the robustness property as is observed in human brain function. Whereas earlier we have shown that AfNNs can still generalize and learn, here we show that these networks are robust against damages occurring after the learning process has terminated. The results support the view that AfNNs embody the important feature of durability. In our contribution, we investigate the durability of the AfNN when some of the neurons in the hidden layer are damaged after the learning process.
ER -