Admissibility of Memorization Learning with Respect to Projection Learning in the Presence of Noise

Akira HIRABAYASHI; Hidemitsu OGAWA; Yukihiko YAMASHITA

Admissibility of Memorization Learning with Respect to Projection Learning in the Presence of Noise

Akira HIRABAYASHI, Hidemitsu OGAWA, Yukihiko YAMASHITA

Full Text Views

0

Cite this

Summary :

In learning of feed-forward neural networks, so-called 'training error' is often minimized. This is, however, not related to the generalization capability which is one of the major goals in the learning. It can be interpreted as a substitute for another learning which considers the generalization capability. Admissibility is a concept to discuss whether a learning can be a substitute for another learning. In this paper, we discuss the case where the learning which minimizes a training error is used as a substitute for the projection learning, which considers the generalization capability, in the presence of noise. Moreover, we give a method for choosing a training set which satisfies the admissibility.

Publication: IEICE TRANSACTIONS on Information Vol.E82-D No.2 pp.488-496

Publication Date: 1999/02/25

Publicized

Online ISSN

DOI

Type of Manuscript: PAPER

Category: Bio-Cybernetics and Neurocomputing

Cite this

Copy

Akira HIRABAYASHI, Hidemitsu OGAWA, Yukihiko YAMASHITA, "Admissibility of Memorization Learning with Respect to Projection Learning in the Presence of Noise" in IEICE TRANSACTIONS on Information, vol. E82-D, no. 2, pp. 488-496, February 1999, doi: .
Abstract: In learning of feed-forward neural networks, so-called 'training error' is often minimized. This is, however, not related to the generalization capability which is one of the major goals in the learning. It can be interpreted as a substitute for another learning which considers the generalization capability. Admissibility is a concept to discuss whether a learning can be a substitute for another learning. In this paper, we discuss the case where the learning which minimizes a training error is used as a substitute for the projection learning, which considers the generalization capability, in the presence of noise. Moreover, we give a method for choosing a training set which satisfies the admissibility.
URL: https://global.ieice.org/en_transactions/information/10.1587/e82-d_2_488/_p

Copy

@ARTICLE{e82-d_2_488,
author={Akira HIRABAYASHI, Hidemitsu OGAWA, Yukihiko YAMASHITA, },
journal={IEICE TRANSACTIONS on Information},
title={Admissibility of Memorization Learning with Respect to Projection Learning in the Presence of Noise},
year={1999},
volume={E82-D},
number={2},
pages={488-496},
abstract={In learning of feed-forward neural networks, so-called 'training error' is often minimized. This is, however, not related to the generalization capability which is one of the major goals in the learning. It can be interpreted as a substitute for another learning which considers the generalization capability. Admissibility is a concept to discuss whether a learning can be a substitute for another learning. In this paper, we discuss the case where the learning which minimizes a training error is used as a substitute for the projection learning, which considers the generalization capability, in the presence of noise. Moreover, we give a method for choosing a training set which satisfies the admissibility.},
keywords={},
doi={},
ISSN={},
month={February},}

Copy

TY - JOUR
TI - Admissibility of Memorization Learning with Respect to Projection Learning in the Presence of Noise
T2 - IEICE TRANSACTIONS on Information
SP - 488
EP - 496
AU - Akira HIRABAYASHI
AU - Hidemitsu OGAWA
AU - Yukihiko YAMASHITA
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E82-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 1999
AB - In learning of feed-forward neural networks, so-called 'training error' is often minimized. This is, however, not related to the generalization capability which is one of the major goals in the learning. It can be interpreted as a substitute for another learning which considers the generalization capability. Admissibility is a concept to discuss whether a learning can be a substitute for another learning. In this paper, we discuss the case where the learning which minimizes a training error is used as a substitute for the projection learning, which considers the generalization capability, in the presence of noise. Moreover, we give a method for choosing a training set which satisfies the admissibility.
ER -