Abstract
In this paper we are describing experiments and results of applications of the continual evolution algorithm to construction and optimization of recurrent neural networks with heterogeneous units. Our algorithm is a hybrid genetic algorithm with sequential individuals replacement, varibale population size and age-based probability control functions. Short introduction to main idea of the algorithm is given. We describe some new features implemented into the algorithm, the encoding of individuals, crossover, and mutation operators. The behavior of population during an evolutionary process is studied on atificial benchmark data sets. Results of the experiments confirm the theoretical properties of the algorithm.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Lampinen Jouni, A., Price Kenneth, V., Storn Rainer, M.: Differential Evolution A Practical Approach to Global Optimization (2005) ISBN 978-3-540-20950-8
Williams, R.J., Zipser, D.: A learning algorithm for continually running fully recurrent neural networks. Neural Computation 1(2), 270–280 (1989)
Williams, R.J.: Gradient-based learning algorithm for recurrent networks and their computational complexity. In: Chauvin, Y., Rumelhart, D.E. (eds.) Back-propagation: Theory, Architectures and Applications (1995)
Yao, X.: Evolving artificial neural networks (1999)
Buk, Z., Šnorek, M.: A brief introduction to the continual evolution algorithm. In: Proceedings of 6th International Conference APLIMAT 2007, Part III, Bratislava, pp. 71–76 (February 2007) ISBN: 978-80-969562-6-5
Buk, Z.: Survey of the computational intelligence methods for data mining. Technical Report DCSE-DTP-2007-01, Czech Technical University in Prague, FEE, CTU Prague, Czech Republic (2007)
Buk, Z., Šnorek, M.: Nature Inspired Methods for Models Creation and Optimization. In: Proceedings of 41th Spring International Conference MOSIS 2007, pp. 56–62. MARQ, Ostrava (2007)
Buk, Z., Šnorek, M.: Continual Evolution Algorithm for Building of ANN-based Models. In: Proceedings of the 6th EUROSIM Congress on Modelling and Simulation, Ljubljana, Slovenia (2007)
De Jong, K.A.: Evolutionary Computation: a unified approach. The MIT Press, Cambridge (2006)
Eiben, A.E., Smith, J.E.: Introduction to Evolutionary Computing. Springer, Heidelberg (2003)
Wong, K.-W., Leung, C.-S., Chang, S.-J.: Handwritten Digit Recognition Using Multi-Layer Feedforward Neural Networks with Periodic and Monotonic Activation Functions. In: 16th International Conference on Pattern Recognition (ICPR 2002). ICPR, vol. 3, p. 30106 (2002)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Buk, Z., Šnorek, M. (2008). Hybrid Evolution of Heterogeneous Neural Networks. In: Kůrková, V., Neruda, R., Koutník, J. (eds) Artificial Neural Networks - ICANN 2008. ICANN 2008. Lecture Notes in Computer Science, vol 5163. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-87536-9_44
Download citation
DOI: https://doi.org/10.1007/978-3-540-87536-9_44
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-87535-2
Online ISBN: 978-3-540-87536-9
eBook Packages: Computer ScienceComputer Science (R0)