gov.sandia.cognition.learning.algorithm.perceptron

Class OnlineMultiPerceptron.ProportionalUpdate<CategoryType>

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.learning.algorithm.AbstractBatchAndIncrementalLearner<InputOutputPair<? extends Vectorizable,CategoryType>,LinearMultiCategorizer<CategoryType>>

gov.sandia.cognition.learning.algorithm.perceptron.OnlineMultiPerceptron<CategoryType>

gov.sandia.cognition.learning.algorithm.perceptron.OnlineMultiPerceptron.ProportionalUpdate<CategoryType>

Type Parameters:

CategoryType - The type of output categories. Can be any type that has a valid equals and hashCode method.

All Implemented Interfaces:

BatchAndIncrementalLearner<InputOutputPair<? extends Vectorizable,CategoryType>,LinearMultiCategorizer<CategoryType>>, BatchLearner<java.util.Collection<? extends InputOutputPair<? extends Vectorizable,CategoryType>>,LinearMultiCategorizer<CategoryType>>, IncrementalLearner<InputOutputPair<? extends Vectorizable,CategoryType>,LinearMultiCategorizer<CategoryType>>, VectorFactoryContainer, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

Enclosing class:

OnlineMultiPerceptron<CategoryType>

@PublicationReference(title="Ultraconservative online algorithms for multiclass problems",
                      author={"Koby Crammer","Yoram Singer"},
                      year=2003,
                      type=Journal,
                      publication="The Journal of Machine Learning Research",
                      pages={951,991},
                      url="http://portal.acm.org/citation.cfm?id=944936")
public static class OnlineMultiPerceptron.ProportionalUpdate<CategoryType>
extends OnlineMultiPerceptron<CategoryType>

Variant of a multi-category Perceptron that performs a proportional weight update on all categories that are scored higher than the true category such that the weights sum to 1.0 and are proportional how much larger the score was for each incorrect category than the true category.

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class gov.sandia.cognition.learning.algorithm.perceptron.OnlineMultiPerceptron

OnlineMultiPerceptron.ProportionalUpdate<CategoryType>, OnlineMultiPerceptron.UniformUpdate<CategoryType>

Field Summary

Fields

Modifier and Type	Field and Description
static double	DEFAULT_MIN_MARGIN The default minimum margin is 0.001.

Fields inherited from class gov.sandia.cognition.learning.algorithm.perceptron.OnlineMultiPerceptron

minMargin, vectorFactory

Constructor Summary

Constructors

Constructor and Description
ProportionalUpdate() Creates a new OnlineMultiPerceptron.ProportionalUpdate.
ProportionalUpdate(double minMargin) Creates a new OnlineMultiPerceptron.ProportionalUpdate with the given minimum margin.
ProportionalUpdate(double minMargin, VectorFactory<?> vectorFactory) Creates a new OnlineMultiPerceptron.ProportionalUpdate with the given minimum margin and backing vector factory.

Method Summary

Modifier and Type	Method and Description
void	setMinMargin(double minMargin) Gets the minimum margin to enforce.
void	update(LinearMultiCategorizer<CategoryType> target, InputOutputPair<? extends Vectorizable,CategoryType> example) The update method updates an object of ResultType using the given new data of type DataType, using some form of "learning" algorithm.

Methods inherited from class gov.sandia.cognition.learning.algorithm.perceptron.OnlineMultiPerceptron

createInitialLearnedObject, getMinMargin, getVectorFactory, setVectorFactory

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

Field Detail

DEFAULT_MIN_MARGIN

public static final double DEFAULT_MIN_MARGIN

The default minimum margin is 0.001.

See Also:

Constant Field Values

Constructor Detail

ProportionalUpdate

public ProportionalUpdate()

Creates a new OnlineMultiPerceptron.ProportionalUpdate.

ProportionalUpdate

public ProportionalUpdate(double minMargin)

Creates a new OnlineMultiPerceptron.ProportionalUpdate with the given minimum margin.

Parameters:

minMargin - The minimum margin to consider an example correct.

ProportionalUpdate

public ProportionalUpdate(double minMargin,
                          VectorFactory<?> vectorFactory)

Creates a new OnlineMultiPerceptron.ProportionalUpdate with the given minimum margin and backing vector factory.

Parameters:

minMargin - The minimum margin to consider an example correct.

vectorFactory - The vector factory used to create the weight vectors.

Method Detail

update

public void update(LinearMultiCategorizer<CategoryType> target,
                   InputOutputPair<? extends Vectorizable,CategoryType> example)

Description copied from interface: IncrementalLearner

The update method updates an object of ResultType using the given new data of type DataType, using some form of "learning" algorithm.

Specified by:

update in interface IncrementalLearner<InputOutputPair<? extends Vectorizable,CategoryType>,LinearMultiCategorizer<CategoryType>>

Overrides:

update in class OnlineMultiPerceptron<CategoryType>

Parameters:

target - The object to update.

example - The new data for the learning algorithm to use to update the object.

setMinMargin

public void setMinMargin(double minMargin)

Description copied from class: OnlineMultiPerceptron

Gets the minimum margin to enforce. Any value less than or equal to this is considered an error for the algorithm.

Overrides:

setMinMargin in class OnlineMultiPerceptron<CategoryType>

Parameters:

minMargin - The minimum margin. Cannot be negative.