Inversion of multilayer nets
- 1 January 1989
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 425-430 vol.2
- https://doi.org/10.1109/ijcnn.1989.118277
Abstract
The method of inversion for arbitrary continuous multilayer nets is developed. The inversion is done by computing iteratively an input vector which minimizes the least-mean-square errors to approximate a given output target. This inversion is not unique for given targets and depends on the starting point in input space. The inversion method turns out to be a valuable tool for the examination of multilayer nets (MLNs). Applications of the inversion method to constraint satisfaction, feature detection, and the testing of reliability and performance of MLNs are outlined. It is concluded that recurrent nets and even time-delay nets might be invertible.Keywords
This publication has 5 references indexed in Scilit:
- Detection of Minimal Microfeatures by Internal FeedbackPublished by Springer Nature ,1989
- Learning to predict by the methods of temporal differencesMachine Learning, 1988
- Backpropagation: past and futurePublished by Institute of Electrical and Electronics Engineers (IEEE) ,1988
- Generalization of back-propagation to recurrent neural networksPhysical Review Letters, 1987
- Learning representations by back-propagating errorsNature, 1986