Module es.upm.fi.cig.multictbnc
Class CTBNHillClimbingHybridAlgorithm
java.lang.Object
es.upm.fi.cig.multictbnc.learning.structure.optimisation.hillclimbing.implementation.CTBNHillClimbingIndividual
es.upm.fi.cig.multictbnc.learning.structure.hybrid.hillclimbing.CTBNHillClimbingHybridAlgorithm
- All Implemented Interfaces:
HillClimbingImplementation
Implements the maximisation phase (hill climbing algorithm) of the hybrid structure learning algorithm for
continuous-time Bayesian networks.
-
Constructor Summary
ConstructorsConstructorDescriptionCTBNHillClimbingHybridAlgorithm(CTBNScoreFunction scoreFunction, boolean[][] initialAdjacencyMatrix) Initialises the algorithm by proving a score function and an initial adjacency matrix for the continuous-time Bayesian network. -
Method Summary
Modifier and TypeMethodDescriptionprotected HillClimbingSolutionfindBestNeighbor(CTBN<? extends Node> ctbn, int idxNode, boolean[][] adjacencyMatrix, Map<Long, Double> cache) Finds the best neighbour for a CTBN node.Methods inherited from class es.upm.fi.cig.multictbnc.learning.structure.optimisation.hillclimbing.implementation.CTBNHillClimbingIndividual
computeScore, findStructure, findStructure, findStructure, findStructureNode, getIdentifier, getInfoScoreFunction, getParametersAlgorithm, increaseNumEdgesTested
-
Constructor Details
-
CTBNHillClimbingHybridAlgorithm
public CTBNHillClimbingHybridAlgorithm(CTBNScoreFunction scoreFunction, boolean[][] initialAdjacencyMatrix) Initialises the algorithm by proving a score function and an initial adjacency matrix for the continuous-time Bayesian network.- Parameters:
scoreFunction- score functioninitialAdjacencyMatrix- adjacency matrix of the initial structure
-
-
Method Details
-
findBestNeighbor
protected HillClimbingSolution findBestNeighbor(CTBN<? extends Node> ctbn, int idxNode, boolean[][] adjacencyMatrix, Map<Long, Double> cache) Description copied from class:CTBNHillClimbingIndividualFinds the best neighbour for a CTBN node.- Overrides:
findBestNeighborin classCTBNHillClimbingIndividual- Parameters:
ctbn- continuous-time Bayesian network that contains the nodeidxNode- node indexadjacencyMatrix- current adjacency matrixcache- cache used to avoid recomputing scores- Returns:
- a
HillClimbingSolutionwith the adjacency matrix and score of the best neighbour
-