On the Activation Function and Fault Tolerance in Feedforward Neural Networks

Nait Charif HAMMADI; Hideo ITO

On the Activation Function and Fault Tolerance in Feedforward Neural Networks

Nait Charif HAMMADI, Hideo ITO

Full Text Views

0

Share
Cite this

Summary :

Considering the pattern classification/recognition tasks, the influence of the activation function on fault tolerance property of feedforward neural networks is empirically investigated. The simulation results show that the activation function largely influences the fault tolerance and the generalization property of neural networks. It is found that, neural networks with symmetric sigmoid activation function are largely fault tolerant than the networks with asymmetric sigmoid function. However the close relation between the fault tolerance and the generalization property was not observed and the networks with asymmetric activation function slightly generalize better than the networks with the symmetric activation function. First, the influence of the activation function on fault tolerance property of neural networks is investigated on the XOR problem, then the results are generalized by evaluating the fault tolerance property of different NNs implementing different benchmark problems.

Publication: IEICE TRANSACTIONS on Information Vol.E81-D No.1 pp.66-72

Publication Date: 1998/01/25

Publicized

Online ISSN

DOI

Type of Manuscript

Category: Fault Tolerant Computing

Cite this

Copy

Nait Charif HAMMADI, Hideo ITO, "On the Activation Function and Fault Tolerance in Feedforward Neural Networks" in IEICE TRANSACTIONS on Information, vol. E81-D, no. 1, pp. 66-72, January 1998, doi: .
Abstract: Considering the pattern classification/recognition tasks, the influence of the activation function on fault tolerance property of feedforward neural networks is empirically investigated. The simulation results show that the activation function largely influences the fault tolerance and the generalization property of neural networks. It is found that, neural networks with symmetric sigmoid activation function are largely fault tolerant than the networks with asymmetric sigmoid function. However the close relation between the fault tolerance and the generalization property was not observed and the networks with asymmetric activation function slightly generalize better than the networks with the symmetric activation function. First, the influence of the activation function on fault tolerance property of neural networks is investigated on the XOR problem, then the results are generalized by evaluating the fault tolerance property of different NNs implementing different benchmark problems.
URL: https://globals.ieice.org/en_transactions/information/10.1587/e81-d_1_66/_p

Copy

@ARTICLE{e81-d_1_66,
author={Nait Charif HAMMADI, Hideo ITO, },
journal={IEICE TRANSACTIONS on Information},
title={On the Activation Function and Fault Tolerance in Feedforward Neural Networks},
year={1998},
volume={E81-D},
number={1},
pages={66-72},
abstract={Considering the pattern classification/recognition tasks, the influence of the activation function on fault tolerance property of feedforward neural networks is empirically investigated. The simulation results show that the activation function largely influences the fault tolerance and the generalization property of neural networks. It is found that, neural networks with symmetric sigmoid activation function are largely fault tolerant than the networks with asymmetric sigmoid function. However the close relation between the fault tolerance and the generalization property was not observed and the networks with asymmetric activation function slightly generalize better than the networks with the symmetric activation function. First, the influence of the activation function on fault tolerance property of neural networks is investigated on the XOR problem, then the results are generalized by evaluating the fault tolerance property of different NNs implementing different benchmark problems.},
keywords={},
doi={},
ISSN={},
month={January},}

Copy

TY - JOUR
TI - On the Activation Function and Fault Tolerance in Feedforward Neural Networks
T2 - IEICE TRANSACTIONS on Information
SP - 66
EP - 72
AU - Nait Charif HAMMADI
AU - Hideo ITO
PY - 1998
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E81-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 1998
AB - Considering the pattern classification/recognition tasks, the influence of the activation function on fault tolerance property of feedforward neural networks is empirically investigated. The simulation results show that the activation function largely influences the fault tolerance and the generalization property of neural networks. It is found that, neural networks with symmetric sigmoid activation function are largely fault tolerant than the networks with asymmetric sigmoid function. However the close relation between the fault tolerance and the generalization property was not observed and the networks with asymmetric activation function slightly generalize better than the networks with the symmetric activation function. First, the influence of the activation function on fault tolerance property of neural networks is investigated on the XOR problem, then the results are generalized by evaluating the fault tolerance property of different NNs implementing different benchmark problems.
ER -