Package ai.libs.jaicore.ml.dyadranking.activelearning

Class ConfidenceIntervalClusteringBasedActiveDyadRanker

java.lang.Object

ai.libs.jaicore.ml.dyadranking.activelearning.ActiveDyadRanker

ai.libs.jaicore.ml.dyadranking.activelearning.ARandomlyInitializingDyadRanker

ai.libs.jaicore.ml.dyadranking.activelearning.ConfidenceIntervalClusteringBasedActiveDyadRanker

public class ConfidenceIntervalClusteringBasedActiveDyadRanker
extends ARandomlyInitializingDyadRanker

A prototypical active dyad ranker based on clustering of pseudo confidence intervals. During the learning procedure, it keeps track over the standard deviation of the skill values predicted for a dyad. First a constant number of random queries is sampled at the beginning. Then the sampling strategy clusteres the skill values of all alternatives for each instance according to the lower and upper bounds of the confidence intervals of the skill for all corresponding dyads. Confidence intervals are given by [skill - std, skill + std] where skill denotes the skill and std denotes the empirical standard deviaition of the skill for a dyad. Afterwards, it picks one of the largest clusters and then selects the two dyads for which the confidence intervals overlap the most within the cluster for pairwise comparison, until a minibatch of constant size is filled.

Field Summary

Fields inherited from class ai.libs.jaicore.ml.dyadranking.activelearning.ActiveDyadRanker

poolProvider, ranker

Constructor Summary

Constructors

Constructor	Description
ConfidenceIntervalClusteringBasedActiveDyadRanker(PLNetDyadRanker ranker, IDyadRankingPoolProvider poolProvider, int seed, int numberRandomQueriesAtStart, int minibatchSize, weka.clusterers.Clusterer clusterer)

Method Summary

Modifier and Type	Method	Description
void	activelyTrainWithOneInstance()

Methods inherited from class ai.libs.jaicore.ml.dyadranking.activelearning.ARandomlyInitializingDyadRanker

activelyTrain, getDyadStats, getInstanceFeatures, getIteration, getMinibatchSize, getNumberRandomQueriesAtStart, getRandom, updateRanker

Methods inherited from class ai.libs.jaicore.ml.dyadranking.activelearning.ActiveDyadRanker

getPoolProvider, getRanker, setPoolProvider, setRanker

Methods inherited from class java.lang.Object

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

Constructor Detail

ConfidenceIntervalClusteringBasedActiveDyadRanker

public ConfidenceIntervalClusteringBasedActiveDyadRanker(PLNetDyadRanker ranker,
                                                         IDyadRankingPoolProvider poolProvider,
                                                         int seed,
                                                         int numberRandomQueriesAtStart,
                                                         int minibatchSize,
                                                         weka.clusterers.Clusterer clusterer)

Method Detail

activelyTrainWithOneInstance

public void activelyTrainWithOneInstance()

Specified by:

activelyTrainWithOneInstance in class ARandomlyInitializingDyadRanker