Acceleration techniques for the network inversion algorithm

Hiroyuki Takizawa; Taira Nakajimat; Masaaki Nishi; Hiroaki Kobayashi; Tadao Nakamura

Acceleration techniques for the network inversion algorithm

Hiroyuki Takizawa, Taira Nakajimat, Masaaki Nishi, Hiroaki Kobayashi, Tadao Nakamura

Tohoku University

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

Abstract

We apply two acceleration techniques for the backpropagation algorithm to an iterative gradient descent algorithm called the network inversion algorithm. Experimental results show that these techniques are also quite effective to decrease the number of iterations required for the detection of input vectors on the classification boundary of a multilayer perceptron.

Original language	English
Pages (from-to)	508-511
Number of pages	4
Journal	IEICE Transactions on Information and Systems
Volume	E82-D
Issue number	2
Publication status	Published - 1999

Keywords

Active learning based on existing training examples
Classification problems
Multilayer perceptrons
The network inversion algorithm

Cite this

@article{ba8f47b62a2f4ce0a69eb4e5cbd49ae5,

title = "Acceleration techniques for the network inversion algorithm",

abstract = "We apply two acceleration techniques for the backpropagation algorithm to an iterative gradient descent algorithm called the network inversion algorithm. Experimental results show that these techniques are also quite effective to decrease the number of iterations required for the detection of input vectors on the classification boundary of a multilayer perceptron.",

keywords = "Active learning based on existing training examples, Classification problems, Multilayer perceptrons, The network inversion algorithm",

author = "Hiroyuki Takizawa and Taira Nakajimat and Masaaki Nishi and Hiroaki Kobayashi and Tadao Nakamura",

year = "1999",

language = "English",

volume = "E82-D",

pages = "508--511",

journal = "IEICE Transactions on Information and Systems",

issn = "0916-8532",

publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",

number = "2",

}

TY - JOUR

T1 - Acceleration techniques for the network inversion algorithm

AU - Takizawa, Hiroyuki

AU - Nakajimat, Taira

AU - Nishi, Masaaki

AU - Kobayashi, Hiroaki

AU - Nakamura, Tadao

PY - 1999

Y1 - 1999

N2 - We apply two acceleration techniques for the backpropagation algorithm to an iterative gradient descent algorithm called the network inversion algorithm. Experimental results show that these techniques are also quite effective to decrease the number of iterations required for the detection of input vectors on the classification boundary of a multilayer perceptron.

AB - We apply two acceleration techniques for the backpropagation algorithm to an iterative gradient descent algorithm called the network inversion algorithm. Experimental results show that these techniques are also quite effective to decrease the number of iterations required for the detection of input vectors on the classification boundary of a multilayer perceptron.

KW - Active learning based on existing training examples

KW - Classification problems

KW - Multilayer perceptrons

KW - The network inversion algorithm

UR - http://www.scopus.com/inward/record.url?scp=0033339877&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033339877&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033339877

SN - 0916-8532

VL - E82-D

SP - 508

EP - 511

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

IS - 2

ER -

Acceleration techniques for the network inversion algorithm

Abstract

Keywords

Other files and links

Fingerprint

Cite this