IMPROVED ADAPTIVE META-NETWORK DESIGN EMPLOYING GENETIC ALGORITHM TECHNIQUES

Ben McElroy, Gareth Howells

Abstract

This paper investigates the employment of a Genetic Algorithm to optimally configure the parameters of a class of weightless artificial neural network architectures. Specifically, the Genetic Algorithm is used to vary the parameters of the architecture and reduce the rigidity of the mutation algorithm to allow for a more varied population and avoidance of local minima traps. An exemplar of the system is presented in the form of an obstacle avoidance system for a mobile robot equipped with ultrasonic sensors.

References

  1. Abraham, A., 2004, Meta learning evolutionary artificial neural networks. Neurocomputing, Issue - 38 : Vol. - 56, 0925-2312.
  2. Abraham A., 2002, Optimization of Evolutionary Neural Networks Using Hybrid Learning Algorithms. International Joint Conference on Neural Networks, 0-7803-7278-6.
  3. Bledsoe W. and Browning I., 1959, Pattern recognition and reading by machine. AFIPS Joint Computer Conferences. Boston, Massachusetts.
  4. Gilbert E., 1958, Gray codes and paths on the n-cube. BellSystem Technical Journal, Vol. 37. 815-826.
  5. Howells G., Fairhurst M. C., and Bisset D. L., 1994. BCN: an architecture for weightless RAM-based neural networks. IEEE International Conference on Neural Networks. Orlando, FL. 0-7803-1901-X.
  6. Howells G., Fairhurst M.C., and Bisset D. L., 1995. GCN: the generalised convergent network. International Conference on Image Processing and its Applications. Edinburgh, 0-85296-642-3.
  7. Kordík, P., Koutník, J, Drchala, J., Kováríka, O., Cepeka, M. and Šnoreka, M., 2010. Meta-learning approach to neural network optimization. Neural Networks, Vol. 23. - 0893-6080.
  8. McElroy B. and Howells G., 2010. Evaluating the Application of Simple Weightless Networks to Complex Patterns International Conference on Emerging Security Technologies (EST), 978-0-7695- 4175-4.
  9. Slowik A. and Bialko M., 2008. Training of artificial neural networks using differential evolution algorithm. Conference on Human System Interactions, 978-1- 4244-1542-7.
  10. Yao, X., 1999 Evolving Artificial Neural Networks. Proceedings of the IEEE, Vol. 87. - 0018-9219 .
Download


Paper Citation


in Harvard Style

McElroy B. and Howells G. (2011). IMPROVED ADAPTIVE META-NETWORK DESIGN EMPLOYING GENETIC ALGORITHM TECHNIQUES . In Proceedings of the 8th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-8425-74-4, pages 142-148. DOI: 10.5220/0003539401420148


in Bibtex Style

@conference{icinco11,
author={Ben McElroy and Gareth Howells},
title={IMPROVED ADAPTIVE META-NETWORK DESIGN EMPLOYING GENETIC ALGORITHM TECHNIQUES},
booktitle={Proceedings of the 8th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2011},
pages={142-148},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003539401420148},
isbn={978-989-8425-74-4},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 8th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - IMPROVED ADAPTIVE META-NETWORK DESIGN EMPLOYING GENETIC ALGORITHM TECHNIQUES
SN - 978-989-8425-74-4
AU - McElroy B.
AU - Howells G.
PY - 2011
SP - 142
EP - 148
DO - 10.5220/0003539401420148