Coloring black boxes: visualization of neural network decisions

  title={Coloring black boxes: visualization of neural network decisions},
  author={Wlodzislaw Duch},
Neural networks are commonly regarded as black boxes performing incomprehensible functions. For classification problems networks provide maps from high dimensional feature space to K-dimensional image space. Images of training vector are projected on polygon vertices, providing visualization of network function. Such visualization may show the dynamics of learning, allow for comparison of different networks, display training vectors around which potential problems may arise, show differences… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 9 times over the past 90 days. VIEW TWEETS


Publications referenced by this paper.
Showing 1-8 of 8 references


  • W. Duch
  • Adamczak and K. Gra̧bczewski, Methodology of…
  • 2001
Highly Influential
4 Excerpts

1994): Neural and statistical methods for the visualization of multidimensional data

  • A. Naud
  • PhD thesis, Dept of Informatics, Nicholaus…
  • 2001
1 Excerpt

NETLAB software

  • I. Nabnay, C. Bishop
  • Aston University, Birmingham, UK
  • 1997
1 Excerpt

Neural networks for pattern recognition

  • C. Bishop
  • Oxford: Clarendon Press
  • 1994
2 Excerpts

Swets , Measuring the accuracy of diagnostic systems

  • A. J.
  • Science
  • 1988

Similar Papers

Loading similar papers…