Module es.upm.fi.cig.multictbnc

Package es.upm.fi.cig.multictbnc.learning.structure.hybrid.hillclimbing

Class BNHillClimbingHybridAlgorithm

java.lang.Object

es.upm.fi.cig.multictbnc.learning.structure.optimisation.hillclimbing.implementation.BNHillClimbing

es.upm.fi.cig.multictbnc.learning.structure.hybrid.hillclimbing.BNHillClimbingHybridAlgorithm

All Implemented Interfaces:

HillClimbingImplementation

public class BNHillClimbingHybridAlgorithm
extends BNHillClimbing

Implements the maximisation phase (hill climbing algorithm) of the hybrid structure learning algorithm for Bayesian networks.

Constructor Summary

Constructors

Constructor	Description
BNHillClimbingHybridAlgorithm(BNScoreFunction scoreFunction, boolean[][] skeletonAdjacencyMatrix)	Initialises the algorithm by proving a score function and a skeleton of the Bayesian network.

Method Summary

Modifier and Type	Method	Description
protected void	findBestNeighbor(BN<? extends Node> bn, HillClimbingSolution bestSolution, double[] scores, boolean[][][] adjacencyMatrices, String operation)	Finds the best neighbour of the adjacency matrix "bestStructure" given an operation to perform (addition, deletion or reversal of arcs).
HillClimbingSolution	findStructure(PGM<? extends Node> pgm)	Finds a structure for a given PGM.

Methods inherited from class es.upm.fi.cig.multictbnc.learning.structure.optimisation.hillclimbing.implementation.BNHillClimbing

computeScore, findStructure, findStructure, getIdentifier, getInfoScoreFunction, getNumEdgesTested, getParametersAlgorithm, increaseNumEdgesTested, isScoreImproved, resetNumEdgesTested

Methods inherited from class java.lang.Object

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

Constructor Details

BNHillClimbingHybridAlgorithm

public BNHillClimbingHybridAlgorithm(BNScoreFunction scoreFunction,
 boolean[][] skeletonAdjacencyMatrix)

Initialises the algorithm by proving a score function and a skeleton of the Bayesian network.

Parameters:

scoreFunction - score function

skeletonAdjacencyMatrix - adjacency matrix of the Bayesian network skeleton

Method Details

findStructure

public HillClimbingSolution findStructure(PGM<? extends Node> pgm)

Description copied from interface: HillClimbingImplementation

Finds a structure for a given PGM.

Specified by:

findStructure in interface HillClimbingImplementation

Overrides:

findStructure in class BNHillClimbing

Parameters:

pgm - a probabilistic graphical model

Returns:

solution given by the hill climbing algorithm

findBestNeighbor

protected void findBestNeighbor(BN<? extends Node> bn,
 HillClimbingSolution bestSolution,
 double[] scores,
 boolean[][][] adjacencyMatrices,
 String operation)

Description copied from class: BNHillClimbing

Finds the best neighbour of the adjacency matrix "bestStructure" given an operation to perform (addition, deletion or reversal of arcs).

Overrides:

findBestNeighbor in class BNHillClimbing

Parameters:

bn - Bayesian network whose structure is being learnt

bestSolution - best structure so far

scores - two-dimensional double array containing the structure scores after applying each of the possible operations over the adjacency matrix

adjacencyMatrices - three-dimensional boolean array containing the resulting adjacency matrices after applying each of the possible operations over the original matrix

operation - operation to perform over the adjacency matrix