A delay damage model selection algorithm for NARX neural networks
- 1 November 1997
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 45 (11), 2719-2730
- https://doi.org/10.1109/78.650098
Abstract
Recurrent neural networks have become popular models for system identification and time series prediction. Nonlinear autoregressive models with exogenous inputs (NARX) neural network models are a popular subclass of recurrent networks and have been used in many applications. Although 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.Keywords
This publication has 44 references indexed in Scilit:
- Pruning algorithms-a surveyIEEE Transactions on Neural Networks, 1993
- The gamma model—A new neural model for temporal processingNeural Networks, 1992
- EmbedologyJournal of Statistical Physics, 1991
- Non-linear system identification using neural networksInternational Journal of Control, 1990
- On the approximate realization of continuous mappings by neural networksNeural Networks, 1989
- Using Relevance to Reduce Network Size AutomaticallyConnection Science, 1989
- Universal coding, information, prediction, and estimationIEEE Transactions on Information Theory, 1984
- The determination of optimum structures for the state space representation of multivariate stochastic processesIEEE Transactions on Automatic Control, 1982
- Geometry from a Time SeriesPhysical Review Letters, 1980
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974