Multi-scale training of a large backpropagation net

E. W. Feiten, O. Martin, S. W. Otto, J. Hutchinson

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


The shortcomings of automatic learning in neural networks can be overcome by incorporating features of the problem into the weight space. We teach a multi-layer perceptron to recognize hand-printed characters by using a multi-scale method. This leads to better performance for learning rates and for generalization than direct use of back-propagation.

Original languageEnglish (US)
Pages (from-to)503-509
Number of pages7
JournalBiological Cybernetics
Issue number6
StatePublished - Apr 1990

All Science Journal Classification (ASJC) codes

  • Biotechnology
  • General Computer Science


Dive into the research topics of 'Multi-scale training of a large backpropagation net'. Together they form a unique fingerprint.

Cite this