com.mwt.explorers

Class VariableActionEpsilonGreedyExplorer<T extends VariableActionContext>

java.lang.Object

com.mwt.explorers.EpsilonGreedyExplorer<T>

com.mwt.explorers.VariableActionEpsilonGreedyExplorer<T>

All Implemented Interfaces:

ConsumePolicy<T>, Explorer<T>

public class VariableActionEpsilonGreedyExplorer<T extends VariableActionContext>
extends EpsilonGreedyExplorer<T>

The epsilon greedy exploration algorithm. This is a good choice if you have no idea which actions should be preferred. Epsilon greedy is also computationally cheap.

Constructor Summary

Constructors

Constructor and Description
VariableActionEpsilonGreedyExplorer(Policy<T> defaultPolicy, float epsilon) The constructor

Method Summary

Modifier and Type	Method and Description
protected int	getNumActions(T context)

Methods inherited from class com.mwt.explorers.EpsilonGreedyExplorer

chooseAction, enableExplore, updatePolicy

Methods inherited from class java.lang.Object

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

Constructor Detail

VariableActionEpsilonGreedyExplorer

public VariableActionEpsilonGreedyExplorer(Policy<T> defaultPolicy,
                                           float epsilon)

The constructor

Parameters:

defaultPolicy - A default function which outputs an action given a context.

epsilon - The probability of a random exploration.

Method Detail

getNumActions

protected int getNumActions(T context)

Overrides:

getNumActions in class EpsilonGreedyExplorer<T extends VariableActionContext>