Input Feature Extraction for Multilayered Perceptrons Using Supervised Principal Component Analysis

@article{Perantonis1999InputFE,
  title={Input Feature Extraction for Multilayered Perceptrons Using Supervised Principal Component Analysis},
  author={Stavros J. Perantonis and Vassilis Virvilis},
  journal={Neural Processing Letters},
  year={1999},
  volume={10},
  pages={243-252}
}
A method is proposed for constructing salient features from a set of features that are given as input to a feedforward neural network used for supervised learning. Combinations of the original features are formed that maximize the sensitivity of the network's outputs with respect to variations of its inputs. The method exhibits some similarity to Principal Component Analysis, but also takes into account supervised character of the learning task. It is applied to classification problems leading… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 17 references

Multilayered Feedforward Networks for Image Segmentation

  • G. Tarr
  • Ph.D. Thesis, Air Force Institute of Technology
  • 1991
Highly Influential
4 Excerpts

Ueber lineare Methoden in der Wahrscheinlichkeitsrechnung

  • K. Karhunen
  • Annales Academiae Scientiarum Fennicae, Series A1…
  • 1947
Highly Influential
4 Excerpts

Characterization of the sonar signals benchmark

  • J.M.T. Moreno, M. B. Gordon
  • Neural Processing Letters, Vol. 7(1), pp. 1-4
  • 1998
1 Excerpt

Adaptive distributed orthogonalization processing for principal components analysis,

  • H. Chen, R.-W. Liu
  • Proc. International Conference on Acoustics…
  • 1992
1 Excerpt

A neural network learning algorithm for adaptive principal component extraction (APEX)

  • S. Y. Kung, K. I. Diamantaras
  • Proc. International Conference on Acoustics…
  • 1990
1 Excerpt

Similar Papers

Loading similar papers…