Abstract
There is investigated the problem of structuring neural networks. This is a crucial task in designing neural networks similar to the corresponding task of designing modular software systems. We propose a particulary new approach of invoking inductive inference techniques for recognizing regularities in the partially given relation to be implemented and for proposing a prestructured net. Furthermore there will be given some mathematical foundations of the choosen approach in form of some convergence theorems by using the back-propagation strategy. Finally we show some connections to the field of PAC-Learning.
This work has been supported by the German Ministry for Research and Technology (BMFT) under grant no. 01 IW 101.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
M.A. Arbib. Brains, Machines, and Mathematics. Springer Verlag, second edition, 1987.
D. Angluin and C.H. Smith. Inductive inference: Theory and methods. Computing Surveys, 15(3):237–269, 1983.
T. Akutsu and A. Takasu. On pac-learnability of functional dependencies. In S. Doshita, K. Furukawa, K.P. Jantke, and T. Nishida, editors, Algorithmic Learning Theory, volume 743 of Lecture Notes in Artifical Intelligence, pages 229–239, 1992.
R. Fagin and M.Y. Vardi. The theory of data dependencies — an overview. In J. Paredeans, editor, Automata, Languages and Programming, volume 172 of Lecture Notes in Computer Science, pages 1–22, 1984.
K.P. Jantke. Algorithmic learning from incomplete information: Principles and problems. In J. Dassow and J. Kelemen, editors, Machines, Languages, and Complexity., volume 381 of Lecture Notes in Computer Science, pages 188–207. Springer Verlag, 1989.
K.P. Jantke and E. Pippig. Inductive inference for structuring neural networks i: The approach. GOSLER Report 01/92, THL, January 1992.
J.S. Judd. Complexity of Connectionist Learning with Various Node Functions. PhD thesis, Dept. Computer and Informations Science, University of Massachusetts, July 1987.
W.S. McCulloch and W. Pitts. A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophysics, 5:115–133, 1943.
E. Pippig. Inductive inference for structuring neural networks ii: First results. GOSLER Report 12/92, THL, July 1992.
D.E. Rumelhart and J.L. McClelland. Parallel Distributed Processing., volume I and II. MIT Press, 1986.
L.G. Valiant. A theory of the leamable. Comm. ACM, 27(11):1134–1142, 1984.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Pippig, E. (1995). Structuring neural networks and PAC-Learning. In: Jantke, K.P., Lange, S. (eds) Algorithmic Learning for Knowledge-Based Systems. Lecture Notes in Computer Science, vol 961. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60217-8_20
Download citation
DOI: https://doi.org/10.1007/3-540-60217-8_20
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60217-0
Online ISBN: 978-3-540-44737-5
eBook Packages: Springer Book Archive