Catalano.MachineLearning.Exploration

Class TabuSearchExploration

java.lang.Object

Catalano.MachineLearning.Exploration.TabuSearchExploration

All Implemented Interfaces:

IExplorationPolicy

public class TabuSearchExploration
extends java.lang.Object
implements IExplorationPolicy

Tabu search exploration policy. The class implements simple tabu search exploration policy, allowing to set certain actions as tabu for a specified amount of iterations. The actual exploration and choosing from non-tabu actions is done by base exploration policy.

Constructor Summary

Constructors

Constructor and Description
TabuSearchExploration(int actions, IExplorationPolicy basePolicy) Initializes a new instance of the TabuSearchExploration

Method Summary

Modifier and Type	Method and Description
int	ChooseAction(double[] actionEstimates) The method chooses an action depending on the provided estimates.
IExplorationPolicy	getBasePolicy() Base exploration policy.
void	ResetTabuList() Reset tabu list.
void	setBasePolicy(IExplorationPolicy basePolicy) Base exploration policy.

Methods inherited from class java.lang.Object

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

Constructor Detail

TabuSearchExploration

public TabuSearchExploration(int actions,
                             IExplorationPolicy basePolicy)

Initializes a new instance of the TabuSearchExploration

Parameters:

actions - Total actions count.

basePolicy - Base exploration policy.

Method Detail

getBasePolicy

public IExplorationPolicy getBasePolicy()

Base exploration policy.

Returns:

Base policy

setBasePolicy

public void setBasePolicy(IExplorationPolicy basePolicy)

Base exploration policy. Base exploration policy is the policy, which is used to choose from non-tabu actions.

Parameters:

basePolicy -

ChooseAction

public int ChooseAction(double[] actionEstimates)

Description copied from interface: IExplorationPolicy

The method chooses an action depending on the provided estimates. the estimates can be any sort of estimate, which values usefulness of the action (expected summary reward, discounted reward, etc).

Specified by:

ChooseAction in interface IExplorationPolicy

Parameters:

actionEstimates - Action estimates.

Returns:

Returns selected action.

ResetTabuList

public void ResetTabuList()

Reset tabu list. Clears tabu list making all actions allowed.