SOLVING NUMBER SERIES - Architectural Properties of Successful Artificial Neural Networks

Marco Ragni, Andreas Klein

2011

Abstract

Any mathematical pattern can be the generation principle for number series. In contrast to most of the application fields of artificial neural networks (ANN) a successful solution does not only require an approximation of the underlying function but to correctly predict the exact next number. We propose a dynamic learning approach and evaluate our method empirically on number series from the Online Encyclopedia of Integer Sequences. Finally, we investigate research questions about the performance of ANNs, structural properties, and the adequate architecture of the ANN to deal successfully with number series.

Paper Citation

in Harvard Style

Ragni M. and Klein A. (2011). SOLVING NUMBER SERIES - Architectural Properties of Successful Artificial Neural Networks . In Proceedings of the International Conference on Neural Computation Theory and Applications - Volume 1: NCTA, (IJCCI 2011) ISBN 978-989-8425-84-3, pages 224-229. DOI: 10.5220/0003682302240229

in Bibtex Style

@conference{ncta11,
author={Marco Ragni and Andreas Klein},
title={SOLVING NUMBER SERIES - Architectural Properties of Successful Artificial Neural Networks},
booktitle={Proceedings of the International Conference on Neural Computation Theory and Applications - Volume 1: NCTA, (IJCCI 2011)},
year={2011},
pages={224-229},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003682302240229},
isbn={978-989-8425-84-3},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Neural Computation Theory and Applications - Volume 1: NCTA, (IJCCI 2011)
TI - SOLVING NUMBER SERIES - Architectural Properties of Successful Artificial Neural Networks
SN - 978-989-8425-84-3
AU - Ragni M.
AU - Klein A.
PY - 2011
SP - 224
EP - 229
DO - 10.5220/0003682302240229