C# Class Accord.MachineLearning.TabuSearchExploration

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.
Inheritance: IExplorationPolicy
ファイルを表示 Open project: accord-net/framework

Public Methods

Method Description
ChooseAction ( double actionEstimates ) : int

Choose an action.

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). The action is choosed from non-tabu actions only.

ResetTabuList ( ) : void

Reset tabu list.

Clears tabu list making all actions allowed.

SetTabuAction ( int action, int tabuTime ) : void

Set tabu action.
TabuSearchExploration ( int actions, IExplorationPolicy basePolicy ) : System

Initializes a new instance of the TabuSearchExploration class.

Method Details

ChooseAction() public method

Choose an action.
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). The action is choosed from non-tabu actions only.
public ChooseAction ( double actionEstimates ) : int
actionEstimates double Action estimates.
return int

ResetTabuList() public method

Reset tabu list.
Clears tabu list making all actions allowed.
public ResetTabuList ( ) : void
return void

SetTabuAction() public method

Set tabu action.
public SetTabuAction ( int action, int tabuTime ) : void
action int Action to set tabu for.
tabuTime int Tabu time in iterations.
return void

TabuSearchExploration() public method

Initializes a new instance of the TabuSearchExploration class.
public TabuSearchExploration ( int actions, IExplorationPolicy basePolicy ) : System
actions int Total actions count.
basePolicy IExplorationPolicy Base exploration policy.
return System