it.uniroma2.sag.kelp.learningalgorithm.budgetedAlgorithm

Class BudgetedLearningAlgorithm

java.lang.Object

it.uniroma2.sag.kelp.learningalgorithm.budgetedAlgorithm.BudgetedLearningAlgorithm

All Implemented Interfaces:

BinaryLearningAlgorithm, LearningAlgorithm, MetaLearningAlgorithm, OnlineLearningAlgorithm

Direct Known Subclasses:

RandomizedBudgetPerceptron, Stoptron

public abstract class BudgetedLearningAlgorithm
extends Object
implements MetaLearningAlgorithm, OnlineLearningAlgorithm, BinaryLearningAlgorithm

It is binary kernel-based online learning method that binds the number of support vector to a fix number (i.e. the budget) Until the budget is not reached the online learning updating policy is the one of the baseAlgorithm that this metaalgorithm is exploiting. When the budget is full, a new updating policy (that must be specified by extending classes) is adopted

Author:

Simone Filice

Field Summary

Fields

Modifier and Type	Field and Description
protected OnlineLearningAlgorithm	baseAlgorithm
protected int	budget
protected Label	label

Constructor Summary

Constructors

Constructor and Description
BudgetedLearningAlgorithm()

Method Summary

Methods

Modifier and Type	Method and Description
OnlineLearningAlgorithm	getBaseAlgorithm() Returns the base algorithm this meta algorithm is based on
int	getBudget() Returns the budget, i.e.
Label	getLabel()
List<Label>	getLabels() Returns the labels representing the concept to be learned.
PredictionFunction	getPredictionFunction() Returns the predictionFunction learned during the training process
void	learn(Dataset dataset) It starts the training process exploiting the provided dataset
Prediction	learn(Example example) Applies the learning process on a single example, updating its current model
abstract Prediction	predictAndLearnWithFullBudget(Example example) Learns from a single example applying a specific policy that must be adopted when the budget is reached
void	setBaseAlgorithm(LearningAlgorithm baseAlgorithm)
void	setBudget(int budget) Sets the budget, i.e.
void	setLabel(Label label)
void	setLabels(List<Label> labels) Sets the labels representing the concept to be learned.

Methods inherited from class java.lang.Object

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

Methods inherited from interface it.uniroma2.sag.kelp.learningalgorithm.LearningAlgorithm

duplicate, reset

Field Detail

budget

protected int budget

baseAlgorithm

protected OnlineLearningAlgorithm baseAlgorithm

label

protected Label label

Constructor Detail

BudgetedLearningAlgorithm

public BudgetedLearningAlgorithm()

Method Detail

getBudget

public int getBudget()

Returns the budget, i.e. the maximum number of support vectors

Returns:

the budget

setBudget

public void setBudget(int budget)

Sets the budget, i.e. the maximum number of support vectors

Parameters:

budget - the budget to set

setBaseAlgorithm

public void setBaseAlgorithm(LearningAlgorithm baseAlgorithm)

Specified by:

setBaseAlgorithm in interface MetaLearningAlgorithm

Parameters:

baseAlgorithm - the baseAlgorithm to set

getBaseAlgorithm

public OnlineLearningAlgorithm getBaseAlgorithm()

Description copied from interface: MetaLearningAlgorithm

Returns the base algorithm this meta algorithm is based on

Specified by:

getBaseAlgorithm in interface MetaLearningAlgorithm

Returns:

the base algorithm

getPredictionFunction

public PredictionFunction getPredictionFunction()

Description copied from interface: LearningAlgorithm

Returns the predictionFunction learned during the training process

Specified by:

getPredictionFunction in interface LearningAlgorithm

Returns:

the predictionFunction learned during the training process

setLabels

public void setLabels(List<Label> labels)

Description copied from interface: LearningAlgorithm

Sets the labels representing the concept to be learned.

Specified by:

setLabels in interface BinaryLearningAlgorithm

Specified by:

setLabels in interface LearningAlgorithm

Parameters:

labels - the labels representing the concept to be learned

getLabels

public List<Label> getLabels()

Description copied from interface: LearningAlgorithm

Returns the labels representing the concept to be learned.

Specified by:

getLabels in interface BinaryLearningAlgorithm

Specified by:

getLabels in interface LearningAlgorithm

Returns:

the labels representing the concept to be learned

learn

public void learn(Dataset dataset)

Description copied from interface: LearningAlgorithm

It starts the training process exploiting the provided dataset

Specified by:

learn in interface LearningAlgorithm

Parameters:

dataset - the training data

learn

public Prediction learn(Example example)

Description copied from interface: OnlineLearningAlgorithm

Applies the learning process on a single example, updating its current model

Specified by:

learn in interface OnlineLearningAlgorithm

Parameters:

example - the instance to be exploited in the learning process

Returns:

the Prediction associated to example before the updating step

predictAndLearnWithFullBudget

public abstract Prediction predictAndLearnWithFullBudget(Example example)

Learns from a single example applying a specific policy that must be adopted when the budget is reached

Parameters:

example - the example to be exploited in the learning process

Returns:

the prediction on the given example, using the old classification function (before any possible update)

getLabel

public Label getLabel()

Specified by:

getLabel in interface BinaryLearningAlgorithm

Returns:

the label associated to the positive class

setLabel

public void setLabel(Label label)

Specified by:

setLabel in interface BinaryLearningAlgorithm