io.improbable.keanu.algorithms.particlefiltering

Class ParticleFilter

java.lang.Object

io.improbable.keanu.algorithms.particlefiltering.ParticleFilter

public class ParticleFilter
extends java.lang.Object

This class allows you to create particle filters to find likely states of a network (i.e. Particles) A particle filtering approach is used to find probable values for the latent vertices of a Bayesian network, given a set of observed vertices. This is done by incrementally increasing the proportion of the graph under consideration, randomly sampling values for newly added latent variables at each stage. Each increment involves the addition of one new observed vertex and the latent vertices that it depend on. This is done for a specified number of 'particles', each of which represents one set of randomly sampled values of the latent vertices in the Bayesian network and has an associated probability. As the proportion of the graph under consideration incrementally grows, less probable particles are culled and more probable particles are duplicated, resulting in a final set of relatively probable particles. This methodology is similar to the Sequential Importance Resampling Algorithm described here (https://www.lancaster.ac.uk/pg/turnerl/PartileFiltering.pdf).

Constructor Summary

Constructors

Constructor and Description
ParticleFilter(java.util.Collection<? extends Vertex> vertices, int numParticles, int resamplingCycles, double resamplingProportion, KeanuRandom random)

Method Summary

Modifier and Type	Method and Description
Particle	getMostProbableParticle()
java.util.List<Particle>	getMostProbableParticles()
java.util.List<Particle>	getSortedMostProbableParticles()
static ParticleFilterBuilder	ofGraph(java.util.Collection<? extends Vertex> vertices)
static ParticleFilterBuilder	ofVertexInGraph(Vertex vertex) Creates a ParticleFilterBuilder by retrieving connected graph from the vertex

Methods inherited from class java.lang.Object

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

Constructor Detail

ParticleFilter

public ParticleFilter(java.util.Collection<? extends Vertex> vertices,
                      int numParticles,
                      int resamplingCycles,
                      double resamplingProportion,
                      KeanuRandom random)

Method Detail

ofVertexInGraph

public static ParticleFilterBuilder ofVertexInGraph(Vertex vertex)

Creates a ParticleFilterBuilder by retrieving connected graph from the vertex

Parameters:

vertex - vertex to retrieve connected graph from

Returns:

this

ofGraph

public static ParticleFilterBuilder ofGraph(java.util.Collection<? extends Vertex> vertices)

getSortedMostProbableParticles

public java.util.List<Particle> getSortedMostProbableParticles()

Returns:

a list of probable particles (network states) sorted in descending order of likelihood

getMostProbableParticle

public Particle getMostProbableParticle()

getMostProbableParticles

public java.util.List<Particle> getMostProbableParticles()