gov.sandia.cognition.learning.algorithm.confidence

Class ConfidenceWeightedDiagonalVariance

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.learning.algorithm.AbstractBatchAndIncrementalLearner<InputOutputPair<? extends InputType,OutputType>,ResultType>

gov.sandia.cognition.learning.algorithm.AbstractSupervisedBatchAndIncrementalLearner<Vectorizable,java.lang.Boolean,DiagonalConfidenceWeightedBinaryCategorizer>

gov.sandia.cognition.learning.algorithm.confidence.ConfidenceWeightedDiagonalVariance

All Implemented Interfaces:

BatchAndIncrementalLearner<InputOutputPair<? extends Vectorizable,java.lang.Boolean>,DiagonalConfidenceWeightedBinaryCategorizer>, BatchLearner<java.util.Collection<? extends InputOutputPair<? extends Vectorizable,java.lang.Boolean>>,DiagonalConfidenceWeightedBinaryCategorizer>, IncrementalLearner<InputOutputPair<? extends Vectorizable,java.lang.Boolean>,DiagonalConfidenceWeightedBinaryCategorizer>, SupervisedBatchAndIncrementalLearner<Vectorizable,java.lang.Boolean,DiagonalConfidenceWeightedBinaryCategorizer>, SupervisedBatchLearner<Vectorizable,java.lang.Boolean,DiagonalConfidenceWeightedBinaryCategorizer>, SupervisedIncrementalLearner<Vectorizable,java.lang.Boolean,DiagonalConfidenceWeightedBinaryCategorizer>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

Direct Known Subclasses:

ConfidenceWeightedDiagonalVarianceProject

@PublicationReference(title="Confidence-Weighted Linear Classification",
                      author={"Mark Dredze","Koby Crammer","Fernando Pereira"},
                      year=2008,
                      type=Conference,
                      publication="International Conference on Machine Learning",
                      url="http://portal.acm.org/citation.cfm?id=1390190")
public class ConfidenceWeightedDiagonalVariance
extends AbstractSupervisedBatchAndIncrementalLearner<Vectorizable,java.lang.Boolean,DiagonalConfidenceWeightedBinaryCategorizer>

An implementation of the Variance algorithm for learning a confidence-weighted linear categorizer. It updates only the diagonal of the covariance matrix, thus computing the variance of each dimension. It is roughly based on the Passive-Aggressive algorithm PA-I, which uses a linear soft margin. This corresponds to the "Variance-drop" version.

Since:

3.3.0

Author:

Justin Basilico

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected double	confidence The confidence to use for updating.
static double	DEFAULT_CONFIDENCE The default confidence is 0.85.
static double	DEFAULT_DEFAULT_VARIANCE The default variance is 1.0.
protected double	defaultVariance The default variance, which the diagonal of the covariance matrix is initialized to.
protected double	phi Phi is the standard score computed from the confidence.

Constructor Summary

Constructors

Constructor and Description
ConfidenceWeightedDiagonalVariance() Creates a new ConfidenceWeightedDiagonalVariance with default parameters.
ConfidenceWeightedDiagonalVariance(double confidence, double defaultVariance) Creates a new ConfidenceWeightedDiagonalVariance with the given parameters.

Method Summary

Modifier and Type	Method and Description
DiagonalConfidenceWeightedBinaryCategorizer	createInitialLearnedObject() Creates a new initial learned object, before any data is given.
double	getConfidence() Gets the confidence to use for updating.
double	getDefaultVariance() Gets the default variance, which the diagonal of the covariance matrix is initialized to.
void	setConfidence(double confidence) Gets the confidence to use for updating.
void	setDefaultVariance(double defaultVariance) Sets the default variance, which the diagonal of the covariance matrix is initialized to.
void	update(DiagonalConfidenceWeightedBinaryCategorizer target, Vector input, boolean label) Updates the target using the given input and associated label.
void	update(DiagonalConfidenceWeightedBinaryCategorizer target, Vectorizable input, java.lang.Boolean output) The update method updates an object of ResultType using the given a new supervised input-output pair, using some form of "learning" algorithm.

Methods inherited from class gov.sandia.cognition.learning.algorithm.AbstractSupervisedBatchAndIncrementalLearner

update

Methods inherited from class gov.sandia.cognition.learning.algorithm.AbstractBatchAndIncrementalLearner

clone, learn, learn, update

Methods inherited from class java.lang.Object

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

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchAndIncrementalLearner

learn

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchLearner

learn

Methods inherited from interface gov.sandia.cognition.learning.algorithm.IncrementalLearner

update

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Field Detail

DEFAULT_CONFIDENCE

public static final double DEFAULT_CONFIDENCE

The default confidence is 0.85.

See Also:

Constant Field Values

DEFAULT_DEFAULT_VARIANCE

public static final double DEFAULT_DEFAULT_VARIANCE

The default variance is 1.0.

See Also:

Constant Field Values

confidence

protected double confidence

The confidence to use for updating. Must be in [0, 1]. Called eta in the paper.

defaultVariance

protected double defaultVariance

The default variance, which the diagonal of the covariance matrix is initialized to. Must be positive. Called a in the paper.

phi

protected double phi

Phi is the standard score computed from the confidence.

Constructor Detail

ConfidenceWeightedDiagonalVariance

public ConfidenceWeightedDiagonalVariance()

Creates a new ConfidenceWeightedDiagonalVariance with default parameters.

ConfidenceWeightedDiagonalVariance

public ConfidenceWeightedDiagonalVariance(double confidence,
                                          double defaultVariance)

Creates a new ConfidenceWeightedDiagonalVariance with the given parameters.

Parameters:

confidence - The confidence to use. Must be in [0, 1].

defaultVariance - The default value to initialize the covariance matrix to.

Method Detail

createInitialLearnedObject

public DiagonalConfidenceWeightedBinaryCategorizer createInitialLearnedObject()

Description copied from interface: IncrementalLearner

Creates a new initial learned object, before any data is given.

Returns:

The initial learned object.

update

public void update(DiagonalConfidenceWeightedBinaryCategorizer target,
                   Vectorizable input,
                   java.lang.Boolean output)

Description copied from interface: SupervisedIncrementalLearner

The update method updates an object of ResultType using the given a new supervised input-output pair, using some form of "learning" algorithm.

Parameters:

target - The object to update.

input - The supervised input to learn from.

output - The supervised output to learn from.

update

public void update(DiagonalConfidenceWeightedBinaryCategorizer target,
                   Vector input,
                   boolean label)

Updates the target using the given input and associated label.

Parameters:

target - The target to update.

input - The supervised input value.

label - The output label associated with the input.

getConfidence

public double getConfidence()

Gets the confidence to use for updating. Must be in [0, 1]. Called eta in the paper.

Returns:

The confidence.

setConfidence

public void setConfidence(double confidence)

Gets the confidence to use for updating. Must be in [0, 1]. Called eta in the paper.

Parameters:

confidence - The confidence. Must be between 0 and 1, inclusive.

getDefaultVariance

public double getDefaultVariance()

Gets the default variance, which the diagonal of the covariance matrix is initialized to. Must be positive. Called a in the paper.

Returns:

The default variance.

setDefaultVariance

public void setDefaultVariance(double defaultVariance)

Sets the default variance, which the diagonal of the covariance matrix is initialized to. Must be positive. Called a in the paper.

Parameters:

defaultVariance - The default variance. Must be positive.