Proving Properties of Neural Networks with Graph Transformations

Fischer I, Koch M, Berthold MR (1998)

Publication Language: English

Publication Type: Conference contribution, Original article

Publication year: 1998

Publisher: IEEE

Edited Volumes: IEEE International Conference on Neural Networks - Conference Proceedings

Book Volume: 1

Pages Range: 457-456

Conference Proceedings Title: Proceedings of the IEEE International Joint Conference on Neural Networks

Event location: Anchorage, AK US

ISBN: 0-7803-4859-1

URI: http://www2.informatik.uni-erlangen.de/publication/download/ijcnn98a.ps.gz

DOI: 10.1109/IJCNN.1998.682307

Abstract

Graph transformations offer a unifying framework to formalize Neural Networks together with their corresponding training algorithms. It is straightforward to describe also topology changing training algorithms with the help of these transformations. One of the benefits using this formal framework is the support for proving properties of the training algorithms. A training algorithm for Probabilistic Neural Networks is used as an example to prove its termination and correctness on the basis of the corresponding graph rewriting rules.

How to cite

APA:

Fischer, I., Koch, M., & Berthold, M.R. (1998). Proving Properties of Neural Networks with Graph Transformations. In Proceedings of the IEEE International Joint Conference on Neural Networks (pp. 457-456). Anchorage, AK, US: IEEE.

MLA:

Fischer, Ingrid, Manuel Koch, and Michael R. Berthold. "Proving Properties of Neural Networks with Graph Transformations." Proceedings of the IEEE International Joint Conference on Neural Networks, Anchorage, AK IEEE, 1998. 457-456.

BibTeX: Download