Abstract
Neural networks using a piecewise linear ramp-step activation function may be interpreted as expressions in the propositional logic of Kleene-Lukasiewicz. These expressions even though information preserving may have a high degree of complexity impairing their understandability. The paper discloses a strategy which combines classical logic with the logic of Lukasiewicz to decompose a complex rule into a set of simpler rules that cover the former.
Work leading to this paper has been partially supported by the Ministry of Education and Research, Germany, under grant BMBF-CH-99/023 and by the Technical University Federico Santa María, Chile, under grant UTFSM/DGIP-Intelligent Data Mining in Complex Systems/2000-2001.
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
Castro J. L., Trillas E.: The Logic of Neural Networks. Mathware and Softcomputing, Vol. V (1), 23–27, (1998)
Giles R.: Lukasiewicz logic and fuzzy set theory. Int. Jr. Man-Machine Studies 8, 313–327, (1976)
Trillas E.: Sobre funciones de negación en la teoría de conjuntos difusos, Stochastica III-I, 47–60, (1979)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Moraga, C., Salinas, L. (2001). Interpreting Neural Networks in the Frame of the Logic of Lukasiewicz. In: Mira, J., Prieto, A. (eds) Connectionist Models of Neurons, Learning Processes, and Artificial Intelligence. IWANN 2001. Lecture Notes in Computer Science, vol 2084. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45720-8_17
Download citation
DOI: https://doi.org/10.1007/3-540-45720-8_17
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42235-8
Online ISBN: 978-3-540-45720-6
eBook Packages: Springer Book Archive