Abstract
In this paper sufficient conditions are derived that ensure that the output of certain Neural Networks may be interpreted as an almost unique probability distribution meaning that any two probability distributions arising as outputs must be close in a sense to be defined. These are rather important in the context of so-called scoring systems arising in a banking environment if one attempts to compute default probabilities. Preliminary experimental evidence is presented showing that these conditions might well apply in practical situations. It is also noted that these conditions may at times prevent good generalization capabilities of the system.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Oxford (1998)
Christianini, N., Shawe-Taylor, J.: An Introduction to Support Vector Machines and other Kernel-Based Learning Methods. Cambridge University Press, Cambridge (2000)
Zentralbank, E.: Die neue Basler Eigenkapitalvereinbarung aus Sicht der EZB. Monatsbericht, Mai (2001) (The New Basel Capital Accord from the Viewpoint of the European Central Bank)
Falkowski, B.-J.: On Scoring Systems with Binary Input Variables. In: Proceedings of the 6th World Multiconference on Systemics, Cybernetics and Informatics, International Institute of Informatics and Systemics, vol. XIII (2002)
Falkowski, B.-J.: Assessing Credit Risk Using a Cost Function. In: Proceedings of the Intl. Conference on Fuzzy Information Processing, vol. II. Tsinghua University Press, Springer (2003)
Gallant, S.I.: Perceptron-based Learning Algorithms. IEEE Transactions on Neural Networks I(2) (1990)
Hand, D.J., Henley, W.E.: Statistical Classification Methods in Consumer Credit Scoring: a Review. J.R. Statist. Soc. A 160(3) (1997)
Haykin, S.: Neural Networks, 2nd edn. Prentice Hall, Englewood Cliffs (1999)
Mendelsohn, S., Smola, A.J. (eds.): Advanced Lectures on Machine Learning. Machine Learning Summer School 2002. LNCS (LNAI), vol. 2660. Springer, Heidelberg (2003)
Shadbolt, J., Taylor, J.G. (eds.): Neural Networks and the Financial Markets. Springer, Heidelberg (2002)
Vapnik, V.N.: Statistical Learning Theory. John Wiley & Sons, Chichester (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Falkowski, BJ. (2004). Interpreting the Output of Certain Neural Networks as Almost Unique Probability. In: Negoita, M.G., Howlett, R.J., Jain, L.C. (eds) Knowledge-Based Intelligent Information and Engineering Systems. KES 2004. Lecture Notes in Computer Science(), vol 3214. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30133-2_89
Download citation
DOI: https://doi.org/10.1007/978-3-540-30133-2_89
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23206-3
Online ISBN: 978-3-540-30133-2
eBook Packages: Springer Book Archive