Abstract
In this paper, we propose a model of forgetting scheme for Neural Network by topologic. In general, parameters architecture is pre-defined for special problem and is not suitable for knowledge learning like human. Based on layered neurons architecture, we gave a ponn algorithm for constructive neural network which has properties of short-long learning procedure. Experimental result has demonstrated that the proposed model can store knowledge in terms of both short term and long term learning process.
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
Galushkin, A.I.: Neural Network Theory. Springer, New York (2007)
Widodo, A., Yang, B.-S.: Mechanical Systems and Signal Processing. Support Vector Machine in Machine Condition Monitoring and Fault Diagnosis 21(6), 2560–2574 (2007)
O’Reilly, R.C.: Biologically Based Computational Models of High-Level Cognition. Science 314(6), 91–94 (2006)
Morrison, J.H., Hof, P.R.: Life and Death of Neurons in the Aging Brain. Science 278(17), 412–419 (1997)
Wang, S.J.: Bionic(topological)pattern recognition-A new model of pattern recognition theory and its applications. Acta Electron. Sinica 30(10), 1–4 (2002)
Benchmark repository-A collection of artificial and real-world data sets, ww.ics.uci.edu/~mlearn/MLSummary.html
Kim, H.K., Rose, R.C.: Cepstrum-domain acoustic feature compensation based on decomposition of speech and noise for ASR in noisy environments. IEEE Trans. Speech Audio Process. 11(5), 435–446 (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhu, S. (2012). A Forgetting Scheme for Layered-Neurons Procedure. In: Lei, J., Wang, F.L., Li, M., Luo, Y. (eds) Network Computing and Information Security. NCIS 2012. Communications in Computer and Information Science, vol 345. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35211-9_45
Download citation
DOI: https://doi.org/10.1007/978-3-642-35211-9_45
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-35210-2
Online ISBN: 978-3-642-35211-9
eBook Packages: Computer ScienceComputer Science (R0)