Past is important: A method for determining memory structure in NARX neural networks

C. Lee Giles; Tsungnan Lin; Bill G. Horne; S. Y. Kung

Past is important: A method for determining memory structure in NARX neural networks

C. Lee Giles, Tsungnan Lin, Bill G. Horne, S. Y. Kung

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Recurrent networks have become popular models for system identification and time series prediction. NARX (Nonlinear AutoRegressive models with eXogenous inputs) network models are a popular subclass of recurrent networks and have been used in many applications. Though embedded memory can be found in all recurrent network models, it is particularly prominent in NARX models. We show that using intelligent memory order selection through pruning and good initial heuristics significantly improves the generalization and predictive performance of these nonlinear systems on problems as diverse as grammatical inference and time series prediction.

Original language	English (US)
Title of host publication	IEEE World Congress on Computational Intelligence
Editors	Anon
Publisher	IEEE
Pages	1834-1839
Number of pages	6
Volume	3
State	Published - 1998
Event	Proceedings of the 1998 IEEE International Joint Conference on Neural Networks. Part 1 (of 3) - Anchorage, AK, USA Duration: May 4 1998 → May 9 1998

Other

Other	Proceedings of the 1998 IEEE International Joint Conference on Neural Networks. Part 1 (of 3)
City	Anchorage, AK, USA
Period	5/4/98 → 5/9/98

All Science Journal Classification (ASJC) codes

Software

Cite this

@inproceedings{3de697557c53431da32909bffa654176,

title = "Past is important: A method for determining memory structure in NARX neural networks",

abstract = "Recurrent networks have become popular models for system identification and time series prediction. NARX (Nonlinear AutoRegressive models with eXogenous inputs) network models are a popular subclass of recurrent networks and have been used in many applications. Though embedded memory can be found in all recurrent network models, it is particularly prominent in NARX models. We show that using intelligent memory order selection through pruning and good initial heuristics significantly improves the generalization and predictive performance of these nonlinear systems on problems as diverse as grammatical inference and time series prediction.",

author = "Giles, {C. Lee} and Tsungnan Lin and Horne, {Bill G.} and Kung, {S. Y.}",

year = "1998",

language = "English (US)",

volume = "3",

pages = "1834--1839",

editor = "Anon",

booktitle = "IEEE World Congress on Computational Intelligence",

publisher = "IEEE",

note = "Proceedings of the 1998 IEEE International Joint Conference on Neural Networks. Part 1 (of 3) ; Conference date: 04-05-1998 Through 09-05-1998",

}

TY - GEN

T1 - Past is important

T2 - Proceedings of the 1998 IEEE International Joint Conference on Neural Networks. Part 1 (of 3)

AU - Giles, C. Lee

AU - Lin, Tsungnan

AU - Horne, Bill G.

AU - Kung, S. Y.

PY - 1998

Y1 - 1998

N2 - Recurrent networks have become popular models for system identification and time series prediction. NARX (Nonlinear AutoRegressive models with eXogenous inputs) network models are a popular subclass of recurrent networks and have been used in many applications. Though embedded memory can be found in all recurrent network models, it is particularly prominent in NARX models. We show that using intelligent memory order selection through pruning and good initial heuristics significantly improves the generalization and predictive performance of these nonlinear systems on problems as diverse as grammatical inference and time series prediction.

AB - Recurrent networks have become popular models for system identification and time series prediction. NARX (Nonlinear AutoRegressive models with eXogenous inputs) network models are a popular subclass of recurrent networks and have been used in many applications. Though embedded memory can be found in all recurrent network models, it is particularly prominent in NARX models. We show that using intelligent memory order selection through pruning and good initial heuristics significantly improves the generalization and predictive performance of these nonlinear systems on problems as diverse as grammatical inference and time series prediction.

UR - http://www.scopus.com/inward/record.url?scp=0031632608&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031632608&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0031632608

VL - 3

SP - 1834

EP - 1839

BT - IEEE World Congress on Computational Intelligence

A2 - Anon, null

PB - IEEE

Y2 - 4 May 1998 through 9 May 1998

ER -

Past is important: A method for determining memory structure in NARX neural networks

Abstract

Other

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this