Multi-Stream Extended Kalman Filter Training of Neural Networks on a SIMD Parallel Machine

Shuhui Li; Donald C. Wunsch; Michael G. Giesselmann; Edgar O'Hair

Multi-Stream Extended Kalman Filter Training of Neural Networks on a SIMD Parallel Machine

Shuhui Li
, Donald C. Wunsch
, Michael G. Giesselmann
, Edgar O'Hair

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

Abstract

The extended Kalman filter (EKF) algorithm has been shown to be advantageous for neural network trainings. This paper presents a method to do the EFK training on a SIMB parallel machine. We use multi-stream decoupled extended Kalman filter (DEKF) training algorithm which can provide more improved trained network weights and efficient use of the parallel resource. The performance of the parallel DEKF training algorithm is studied and simulation results for the estimation of the wind power using neural networks are provided.

Original language	American English
Journal	Intelligent Engineering Systems Through Artificial Neural Networks
State	Published - Nov 10 1999

Keywords

Artificial Intelligence
Neural Networks

Disciplines

Electrical and Computer Engineering

Cite this

@article{dbcbe2a9a5134b42812962c72a027251,

title = "Multi-Stream Extended Kalman Filter Training of Neural Networks on a SIMD Parallel Machine",

abstract = " The extended Kalman filter (EKF) algorithm has been shown to be advantageous for neural network trainings. This paper presents a method to do the EFK training on a SIMB parallel machine. We use multi-stream decoupled extended Kalman filter (DEKF) training algorithm which can provide more improved trained network weights and efficient use of the parallel resource. The performance of the parallel DEKF training algorithm is studied and simulation results for the estimation of the wind power using neural networks are provided.",

keywords = "Artificial Intelligence, Neural Networks",

author = "Shuhui Li and Wunsch, \{Donald C.\} and Giesselmann, \{Michael G.\} and Edgar O'Hair",

year = "1999",

month = nov,

day = "10",

language = "American English",

journal = "Intelligent Engineering Systems Through Artificial Neural Networks",

}

TY - JOUR

T1 - Multi-Stream Extended Kalman Filter Training of Neural Networks on a SIMD Parallel Machine

AU - Li, Shuhui

AU - Wunsch, Donald C.

AU - Giesselmann, Michael G.

AU - O'Hair, Edgar

PY - 1999/11/10

Y1 - 1999/11/10

N2 - The extended Kalman filter (EKF) algorithm has been shown to be advantageous for neural network trainings. This paper presents a method to do the EFK training on a SIMB parallel machine. We use multi-stream decoupled extended Kalman filter (DEKF) training algorithm which can provide more improved trained network weights and efficient use of the parallel resource. The performance of the parallel DEKF training algorithm is studied and simulation results for the estimation of the wind power using neural networks are provided.

AB - The extended Kalman filter (EKF) algorithm has been shown to be advantageous for neural network trainings. This paper presents a method to do the EFK training on a SIMB parallel machine. We use multi-stream decoupled extended Kalman filter (DEKF) training algorithm which can provide more improved trained network weights and efficient use of the parallel resource. The performance of the parallel DEKF training algorithm is studied and simulation results for the estimation of the wind power using neural networks are provided.

KW - Artificial Intelligence

KW - Neural Networks

UR - https://scholarsmine.mst.edu/ele_comeng_facwork/152

M3 - Article

JO - Intelligent Engineering Systems Through Artificial Neural Networks

JF - Intelligent Engineering Systems Through Artificial Neural Networks

ER -

Multi-Stream Extended Kalman Filter Training of Neural Networks on a SIMD Parallel Machine

Abstract

Keywords

Disciplines

Other files and links

Fingerprint

Cite this