Abstract
By the application of a technique from the statistical pattern recognition literature, the editing algorithm [Devijver 1982], a learning set can be transformed to a data set in which overlap between classes is effectively removed. Because it can be proven that an edited data set is close to Bayes-optimal for the nearest-neighbour classifier, it is very likely that a multi-layer network which classifies all samples in the edited learning set correctly, is also close to Bayes-optimal. In this paper we investigate the performance of the backpropagation algorithm on edited data sets. This leads to an optimal criterion to stop the learning phase and to a moderate improvement in learning speed.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
