Package com.bayesserver.learning.parameters

Class OnlineLearning

java.lang.Object

com.bayesserver.learning.parameters.OnlineLearning

public final class OnlineLearning
extends Object

Adapts the parameters of a Bayesian network, using Bayesian statistics.

Constructor Summary

Constructors

Constructor	Description
OnlineLearning(Network network, InferenceFactory factory)	Initializes a new instance of the OnlineLearning class.

Method Summary

Modifier and Type	Method	Description
void	adapt(Evidence evidence, OnlineLearningOptions options)	Adapt the parameters of a Bayesian network using Bayesian statistics.
Evidence	getEvidence()	Gets the evidence used internally.

Methods inherited from class java.lang.Object

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

Constructor Detail

OnlineLearning

public OnlineLearning(Network network,
                      InferenceFactory factory)

Initializes a new instance of the OnlineLearning class. Learning uses inference as a subroutine, and creates one or more inference engines via the [factory] parameter.

Parameters:

network - The network whose parameters are being adapted.

factory - The inference factory used to create inference engines in cases when learning requires inference.

Method Detail

getEvidence

public Evidence getEvidence()

Gets the evidence used internally. Setting evidence on this instance, and passing it to Adapt saves a copy.

adapt

public void adapt(Evidence evidence,
                  OnlineLearningOptions options)
           throws InconsistentEvidenceException

Adapt the parameters of a Bayesian network using Bayesian statistics.

For nodes to be adapted, they must have Experience tables assigned (and optionally fading tables).

In the case a discrete node, the experience table combined with the probability are used to create a Dirichlet distribution. This distribution acts as a prior during the Bayesian inference process.

Parameters:

evidence - The evidence to learn.

options - Options that affect how parameters are adapted.

Throws:

InconsistentEvidenceException