A Systematic Approach to the Optimal Design of Feed Forward Neural Networks Applied to Log-Synthesis

Rudolf Konrad Fruhwirth; Sepp Peter Steinlechner

A Systematic Approach to the Optimal Design of Feed Forward Neural Networks Applied to Log-Synthesis

Rudolf Konrad Fruhwirth
, Sepp Peter Steinlechner

Chair of Geoenergy Production Engineering (550)

Research output: Contribution to conference › Paper › peer-review

Abstract

G001 A SYSTEMATIC APPROACH TO THE OPTIMAL DESIGN OF FEED FORWARD NEURAL NETWORKS APPLIED TO LOG-SYNTHESIS Abstract 1 Neural networks are increasingly used in geophysical applications. Optimizing neural networks is still a matter of experience and trial and error where network initialization and network size are the most challenging issues. We expanded conventional learning rules to a completely forward connected network including input neurons for automatic normalization of the data. In addition we developed a method for the network initialization based on the statistical properties of the input and output data generating an initial network state that ascertains a fast

Original language	English
Publication status	Published - 2004
Event	66th EAGE Conference & Exhibition 2004 - Paris, France Duration: 7 Jun 2004 → 10 Jun 2004

Conference

Conference	66th EAGE Conference & Exhibition 2004
Country/Territory	France
City	Paris
Period	7/06/04 → 10/06/04

Cite this

@conference{5979484768864f4b827f1fb67a9fc2f3,

title = "A Systematic Approach to the Optimal Design of Feed Forward Neural Networks Applied to Log-Synthesis",

abstract = "G001 A SYSTEMATIC APPROACH TO THE OPTIMAL DESIGN OF FEED FORWARD NEURAL NETWORKS APPLIED TO LOG-SYNTHESIS Abstract 1 Neural networks are increasingly used in geophysical applications. Optimizing neural networks is still a matter of experience and trial and error where network initialization and network size are the most challenging issues. We expanded conventional learning rules to a completely forward connected network including input neurons for automatic normalization of the data. In addition we developed a method for the network initialization based on the statistical properties of the input and output data generating an initial network state that ascertains a fast",

author = "Fruhwirth, \{Rudolf Konrad\} and Steinlechner, \{Sepp Peter\}",

year = "2004",

language = "English",

note = "66th EAGE Conference \& Exhibition 2004 ; Conference date: 07-06-2004 Through 10-06-2004",

}

TY - CONF

T1 - A Systematic Approach to the Optimal Design of Feed Forward Neural Networks Applied to Log-Synthesis

AU - Fruhwirth, Rudolf Konrad

AU - Steinlechner, Sepp Peter

PY - 2004

Y1 - 2004

N2 - G001 A SYSTEMATIC APPROACH TO THE OPTIMAL DESIGN OF FEED FORWARD NEURAL NETWORKS APPLIED TO LOG-SYNTHESIS Abstract 1 Neural networks are increasingly used in geophysical applications. Optimizing neural networks is still a matter of experience and trial and error where network initialization and network size are the most challenging issues. We expanded conventional learning rules to a completely forward connected network including input neurons for automatic normalization of the data. In addition we developed a method for the network initialization based on the statistical properties of the input and output data generating an initial network state that ascertains a fast

AB - G001 A SYSTEMATIC APPROACH TO THE OPTIMAL DESIGN OF FEED FORWARD NEURAL NETWORKS APPLIED TO LOG-SYNTHESIS Abstract 1 Neural networks are increasingly used in geophysical applications. Optimizing neural networks is still a matter of experience and trial and error where network initialization and network size are the most challenging issues. We expanded conventional learning rules to a completely forward connected network including input neurons for automatic normalization of the data. In addition we developed a method for the network initialization based on the statistical properties of the input and output data generating an initial network state that ascertains a fast

M3 - Paper

T2 - 66th EAGE Conference & Exhibition 2004

Y2 - 7 June 2004 through 10 June 2004

ER -