Including Metric Space Topology in Neural Networks Training by Ordering Patterns

Cezary Dendek, Jacek Mandziuk

Abstract

In this paper a new approach to the problem of ordering data in neural network training is presented. According to conducted research, generalization error visibly depends on the order of the training examples. Construction of an order gives some possibility to incorporate knowledge about structure of input and output space into the training process. Simulation results conducted for the isolated handwritten digit recognition problem confirmed the above claims.
BibTeX
@inproceedings{dendek2006including,
  author = {Cezary Dendek and Jacek Mandziuk},
  title = {{Including Metric Space Topology in Neural Networks Training by Ordering Patterns}},
  booktitle = {ICANN},
  year = {2006},
  doi = {10.1007/11840930\_67},
}

Loading related content...

Figure