Adaptive Formation of Optimal Associative Mappings

  • Teuvo Kohonen
Part of the Communication and Cybernetics book series (COMMUNICATION, volume 17)


The idea of machine learning was originally based on the application of variable elements which change their parametric values by the signals they transmit; later, learning processes have been discussed on a more abstract algorithmic level, and mathematical statistics has become the general setting of these problems. Nevertheless, in its genuine form machine learning is a physical process.


Input Vector Input Pattern Projection Matrix Adaptive Filter Stochastic Approximation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 1977

Authors and Affiliations

  • Teuvo Kohonen
    • 1
  1. 1.Dept. of Technical PhysicsHelsinki University of TechnologyOtaniemiFinland

Personalised recommendations