org.joone.engine

Class KohonenSynapse

java.lang.Object

org.joone.engine.Synapse

org.joone.engine.FullSynapse

org.joone.engine.KohonenSynapse

All Implemented Interfaces:

java.io.Serializable, java.util.EventListener, InputPatternListener, Learnable, LearnableSynapse, NeuralElement, NeuralNetListener, OutputPatternListener, Inspectable

public class KohonenSynapse
extends FullSynapse
implements NeuralNetListener

This is an unsupervised Kohonen Synapse which is a Self Organising Map.

This KohonenSynapse works in conjunction with the next layer which should implement a SOM strategy such as a GuassianLayer or WTALayer (Winner Takes All). This synapse should connect to one of these layers, without a SOM Strategy in the next layer this component will not function correctly.

This KohonenSynapse takes a pattern from the previous layer, calculates the distance between the input vector and the weights and passes this on to the next layer. In the learning phase the next layer should calculate the distance fall off between the winner and all other nodes (1.0 being the closest distance and 0.0 being furthest away and not being considered near to the winner). These distances are passed back to this KohonenSynapse and used to adjust the weights.

The weights are adjusted based on the current learning rate and distance fall off.

At each epoch/cycel the learning rate is adjusted in the following way ...

If the current cycle is within the ordering phase then the learning rate is set to

User setup learning rate * exp(-(double)(Current Cycle/Time Constant)).

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
(package private) double	currentLearningRate

Fields inherited from class org.joone.engine.Synapse

array, b_pattern, bitems, bouts, count, enabled, fwdLock, inps, items, learnable, m_batch, m_pattern, myLearner, outs, revLock

Constructor Summary

Constructors

Constructor and Description
KohonenSynapse() The default constructor for the KohonenSynapse class.

Method Summary

Methods

Modifier and Type	Method and Description
protected void	backward(double[] pattern) Adjusts the weights of this Kohonen Synapse according to the neighborhood fall off distance calculated by the next layer.
java.util.TreeSet	check() Check that there are no errors or problems with the properties of this KohonenSynapse.
void	cicleTerminated(NeuralNetEvent e) Changes the learning rate for this synapse depending in the current epoch number.
void	errorChanged(NeuralNetEvent e) Not implemented.
protected void	forward(double[] pattern) Fowards the euclidean distance squared between the input vector and the weight vector to the next layer.
Learner	getLearner() Deprecated. - Used only for backward compatibility
int	getOrderingPhase() Getter for property orderingPhase.
double	getTimeConstant() Getter for property timeConstant.
void	netStarted(NeuralNetEvent e) Initialises any shape sizes such as circular radius and time constant before possible training.
void	netStopped(NeuralNetEvent e) Not implemented.
void	netStoppedError(NeuralNetEvent e, java.lang.String error) Not implemented.
void	setMonitor(Monitor newMonitor) Sets the Monitor object of the synapse
void	setOrderingPhase(int orderingPhase) Setter for property orderingPhase.
void	setTimeConstant(double timeConstant) Setter for property timeConstant.

Methods inherited from class org.joone.engine.FullSynapse

setArrays, setDimensions

Methods inherited from class org.joone.engine.Synapse

addNoise, canCountSteps, fwdGet, fwdPattern_consume, fwdPattern_produce, fwdPut, getCount, getFwdLock, getIgnoreBefore, getInputDimension, getLearningRate, getMomentum, getMonitor, getName, getOutputDimension, getRevLock, getSynapseNameSafely, getWeights, init, initLearner, InspectableTitle, Inspections, isEnabled, isInputFull, isLoopBack, isOutputFull, randomize, readResolve, reset, revGet, revPattern_consume, revPattern_produce, revPut, setEnabled, setIgnoreBefore, setInputDimension, setInputFull, setLearningRate, setLoopBack, setMomentum, setName, setOutputDimension, setOutputFull, setWeights, warnLogger, warnLogger

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.joone.engine.LearnableSynapse

getInputDimension, getLearningRate, getMomentum, getOutputDimension, getWeights, setWeights

Methods inherited from interface org.joone.engine.Learnable

getMonitor, initLearner

Field Detail

currentLearningRate

double currentLearningRate

Constructor Detail

KohonenSynapse

public KohonenSynapse()

The default constructor for the KohonenSynapse class.

Method Detail

backward

protected void backward(double[] pattern)

Adjusts the weights of this Kohonen Synapse according to the neighborhood fall off distance calculated by the next layer.

Overrides:

backward in class FullSynapse

Parameters:

pattern - The pattern with the distance fall off's between the winner and all other nodes. (1.0 is the winner through 0.0 having no similarity to the original input vector)

forward

protected void forward(double[] pattern)

Fowards the euclidean distance squared between the input vector and the weight vector to the next layer. If the learning phase is currently active then the next layer should process this and pass back the distance fall off between the winning output and all other outputs.

Overrides:

forward in class FullSynapse

Parameters:

pattern - The pattern containg the euclidean distance squared between each weight and the input.

setMonitor

public void setMonitor(Monitor newMonitor)

Sets the Monitor object of the synapse

Specified by:

setMonitor in interface NeuralElement

Overrides:

setMonitor in class Synapse

Parameters:

newMonitor - neural.engine.Monitor

cicleTerminated

public void cicleTerminated(NeuralNetEvent e)

Changes the learning rate for this synapse depending in the current epoch number. The learning rate is changed in the following way ...

User setup learning rate * exp(-(double)(Current Cycle/Time Constant)).

Specified by:

cicleTerminated in interface NeuralNetListener

Parameters:

e - The original Net Event.

errorChanged

public void errorChanged(NeuralNetEvent e)

Not implemented.

Specified by:

errorChanged in interface NeuralNetListener

Parameters:

e - The original Net Event.

netStarted

public void netStarted(NeuralNetEvent e)

Initialises any shape sizes such as circular radius and time constant before possible training.

Specified by:

netStarted in interface NeuralNetListener

Parameters:

e - The original Net Event.

netStopped

public void netStopped(NeuralNetEvent e)

Not implemented.

Specified by:

netStopped in interface NeuralNetListener

Parameters:

e - The original Net Event.

netStoppedError

public void netStoppedError(NeuralNetEvent e,
                   java.lang.String error)

Not implemented.

Specified by:

netStoppedError in interface NeuralNetListener

Parameters:

e - The original Net Event.

error - The error that caused this NetStoppedError event.

check

public java.util.TreeSet check()

Check that there are no errors or problems with the properties of this KohonenSynapse.

Specified by:

check in interface NeuralElement

Overrides:

check in class Synapse

Returns:

The TreeSet of errors / problems if any.

See Also:

InputPaternListener, OutputPaternListener

getOrderingPhase

public int getOrderingPhase()

Getter for property orderingPhase.

Returns:

Value of property orderingPhase.

setOrderingPhase

public void setOrderingPhase(int orderingPhase)

Setter for property orderingPhase.

Parameters:

orderingPhase - New value of property orderingPhase.

getTimeConstant

public double getTimeConstant()

Getter for property timeConstant.

Returns:

Value of property timeConstant.

setTimeConstant

public void setTimeConstant(double timeConstant)

Setter for property timeConstant.

Parameters:

timeConstant - New value of property timeConstant.

getLearner

public Learner getLearner()

Deprecated. - Used only for backward compatibility

Description copied from class: Synapse

Returns the appropriate Learner object for this class depending on the Monitor.learningMode property value

Specified by:

getLearner in interface Learnable

Overrides:

getLearner in class FullSynapse

Returns:

the Learner object if applicable, otherwise null

See Also:

Learnable.getLearner()