com.mapr.stats.bandit

Class BetaBayesModel

java.lang.Object

com.mapr.stats.bandit.BayesianBandit

com.mapr.stats.bandit.BetaBayesModel

All Implemented Interfaces:

Iterable<AbstractBayesianDistribution>

public class BetaBayesModel
extends BayesianBandit

Multi-armed bandit problem where each probability is modeled by a beta prior and data about positive and negative trials. An arm is selected by sampling from the current posterior for each arm and picking the one with higher sampled probability.

Constructor Summary

Constructors

Constructor and Description
BetaBayesModel()
BetaBayesModel(int bandits, Random gen)

Method Summary

Methods inherited from class com.mapr.stats.bandit.BayesianBandit

addModelDistribution, getMean, iterator, rank, sample, train

Methods inherited from class java.lang.Object

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

Constructor Detail

BetaBayesModel

public BetaBayesModel()

BetaBayesModel

public BetaBayesModel(int bandits,
              Random gen)