aima.learning.reinforcement
Class QTable<STATE_TYPE,ACTION_TYPE>

java.lang.Object
  aima.learning.reinforcement.QTable<STATE_TYPE,ACTION_TYPE>

public class QTable<STATE_TYPE,ACTION_TYPE>
extends java.lang.Object

Constructor Summary

QTable(java.util.List<ACTION_TYPE> allPossibleActions)

Method Summary

MDPPolicy<STATE_TYPE,ACTION_TYPE>	getPolicy()
java.lang.Double	getQValue(STATE_TYPE state, ACTION_TYPE action)
Pair<ACTION_TYPE,java.lang.Double>	maxDiff(STATE_TYPE startState, ACTION_TYPE action, STATE_TYPE endState)
void	normalize()
void	setQValue(STATE_TYPE state, ACTION_TYPE action, java.lang.Double d)
java.lang.String	toString()
ACTION_TYPE	upDateQ(STATE_TYPE startState, ACTION_TYPE action, STATE_TYPE endState, double alpha, double reward, double phi)

Methods inherited from class java.lang.Object

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

Constructor Detail

QTable

public QTable(java.util.List<ACTION_TYPE> allPossibleActions)

Method Detail

getQValue

public java.lang.Double getQValue(STATE_TYPE state,
                                  ACTION_TYPE action)

maxDiff

public Pair<ACTION_TYPE,java.lang.Double> maxDiff(STATE_TYPE startState,
                                                  ACTION_TYPE action,
                                                  STATE_TYPE endState)

setQValue

public void setQValue(STATE_TYPE state,
                      ACTION_TYPE action,
                      java.lang.Double d)

upDateQ

public ACTION_TYPE upDateQ(STATE_TYPE startState,
                           ACTION_TYPE action,
                           STATE_TYPE endState,
                           double alpha,
                           double reward,
                           double phi)

normalize

public void normalize()

getPolicy

public MDPPolicy<STATE_TYPE,ACTION_TYPE> getPolicy()

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object