Dynamic subgrouping in RTRL provides a faster O(N/sup 2/) algorithm

Static grouping of processing elements (PEs) has been proposed to reduce the computational complexity of real time recurrent learning (RTRL) from O(n/sup 4/) to O(n/sup 2/), but performance suffers. This paper proposes a dynamic subgrouping of PEs estimated from a local approximation of the /spl pi/...

Full description

Saved in:
Bibliographic Details
Published in2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100) Vol. 6; pp. 3418 - 3421 vol.6
Main Authors Euliano, N.R., Principe, J.C.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2000
Subjects
Computational complexity
Computer networks
Feedforward neural networks
History
Laboratories
Neural engineering
Neural networks
Recurrent neural networks
Sun
System identification
Online AccessGet full text
ISBN9780780362932
0780362934
ISSN1520-6149
DOI10.1109/ICASSP.2000.860135

Cover

More Information
Summary:Static grouping of processing elements (PEs) has been proposed to reduce the computational complexity of real time recurrent learning (RTRL) from O(n/sup 4/) to O(n/sup 2/), but performance suffers. This paper proposes a dynamic subgrouping of PEs estimated from a local approximation of the /spl pi/ matrix based on temporal Hebbian of sensitivities during training. The method is O(n/sup 2/) and leads to better performance.
ISBN:9780780362932
0780362934
ISSN:1520-6149
DOI:10.1109/ICASSP.2000.860135