Abstract
In this paper, we address the difficult problem of the learning verification in multilayer neural networks. Finding the activation/inhibition power of each input feature allows us to build synthetic examples and then to find out the minimal recognized patterns as long as to evaluate the robustness of the system. A small illustration upon character recognition clearly shows the interest in bias reduction. A real world application of transients recognition in underwater acoustic helped us to build more efficient features and to significantly improve the generalization rates.
This work has been granted by DCN Ingénierie Sud/LSM under contract C 95 50 638 000
Preview
Unable to display preview. Download preview PDF.
References
Ezhov A.A. & Vvedensky V.L.: Object generation with neural networks. Neural Networks, vol.9, no9, pp.1491–1495, dec96.
Fechner T. & Hinze A.: Delta Analysis: a method for the determination of input feature significance in neural networks. NeuroNîmes'93, pp.393–398, Nimes, France.
Nocera P. & Quélavoine R.: Diminishing the number of nodes in multilayered neural networks. ICNN'94, vol.7, pp.4421–4424, 28jun–2jul94, Orlando, FL.
Oppizzi O. & Quélavoine R.: Rescoring under fuzzy degrees with a multilayered neural network in a rule-based speech recognition system. ICASSP'97, 21–24apr97, Munchen, Germany.
Quélavoine R., Nocera P. & Di Martino M.: Multilayered neural networks and errors in learning corpus. WCNN'96, pp.287–290, 16–18sep96, San Diego, CA.
Quélavoine R.: Étude de l'apprentissage et des structures des rseaux de neurones multicouches pour l'analyse de données.. Thesis, 116p., 13jan97, University of Avignon, France.
Yoda M., Baba K. & Enbutu I.: Explicit representation of knowledge acquired from plant historical data using neural networks IJCNN'91, vol.3, pp.155–160, San Diego, CA.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Quélavoine, R., Nocera, P. (1997). Learning verification in multilayer neural networks. In: Gerstner, W., Germond, A., Hasler, M., Nicoud, JD. (eds) Artificial Neural Networks — ICANN'97. ICANN 1997. Lecture Notes in Computer Science, vol 1327. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020202
Download citation
DOI: https://doi.org/10.1007/BFb0020202
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63631-1
Online ISBN: 978-3-540-69620-9
eBook Packages: Springer Book Archive