predictability minimization

[Juergen Schmidhuber]

  Processing Images by Semi-Linear Predictability Minimization 
  ------------------------------------------------------------
  Nicol N. Schraudolph, Martin Eldracher & Juergen Schmidhuber 
  IDSIA, Lugano, Switzerland                      www.idsia.ch
  Network: Computation in Neural Systems,  10(2):133-169, 1999

In the predictability minimization approach (Neural Computation,
4(6):863-879, 1992), input patterns are fed into a system consisting
of adaptive, initially unstructured feature detectors.   There are also
adaptive predictors constantly trying to predict current feature detector
outputs from other feature detector outputs. Simultaneously, however,
the feature detectors try to become as unpredictable as possible,
resulting in a co-evolution of predictors and feature detectors.
This paper describes the implementation of a visual processing system
trained by  semi-linear  predictability  minimization, and presents many
experiments that examine its response to artificial and real-world images.
In particular, we observe that under a wide variety of conditions,
predictability minimization results in the development of well-known
visual feature detectors.

ftp://ftp.idsia.ch/pub/juergen/pm.ps.gz

Many additional papers now also available in postscript format:
http://www.idsia.ch/~juergen/onlinepub.html

Nici & Martin & Juergen