org.encog.ml
Interface MLAutoAssocation

All Superinterfaces:
MLInput, MLInputOutput, MLMethod, MLOutput, MLRegression
All Known Implementing Classes:
BoltzmannMachine, HopfieldNetwork, ThermalNetwork

public interface MLAutoAssocation
extends MLRegression

Defines a MLMethod that can handle autoassocation. Autoassociation is a simple form of pattern recognition where the MLMethod echos back the exact pattern that the input most closely matches. For example, if the autoassociative MLMethod were trained to recognize an 8x8 grid of characters, the return value would be the entire 8x8 grid of the character recognized. This is the type of recognition performed by Hopfield Networks. It is also an optional recognition form used by GR/PNN's. This is a form of unsupervised training.


Method Summary
 
Methods inherited from interface org.encog.ml.MLRegression
compute
 
Methods inherited from interface org.encog.ml.MLInput
getInputCount
 
Methods inherited from interface org.encog.ml.MLOutput
getOutputCount
 



Copyright © 2014. All Rights Reserved.