public double GetReward(QQuadrant quadrant)
{
var list = GetPossibleActions(quadrant);
double actionReward = list.Max(a => a.Reward);
return(quadrant.Reward + (0.5 * actionReward));
}
private static QQuadrant[,] BuildMapMatrix()
{
var aux = new string[5] {
"A", "B", "C", "D", "E"
};
var matrix = new QQuadrant[5, 10];
for (int i = 0; i < 5; i++)
{
for (int j = 0; j < 10; j++)
{
matrix[i, j] = new QQuadrant(-1, aux[i] + j);
}
}
matrix[3, 1].Reward = -100;
matrix[3, 2].Reward = -100;
matrix[3, 4].Reward = -100;
matrix[3, 5].Reward = -100;
matrix[3, 7].Reward = -100;
matrix[1, 3].Reward = -100;
matrix[1, 7].Reward = -100;
for (int i = 0; i < 9; i++)
{
matrix[0, i].Reward = -100;
}
matrix[0, 9].Reward = 100;
return(matrix);
}
public QAction GetNextAction(QQuadrant quadrant, bool includeRandomization = true)
{
var possibleActions = GetPossibleActions(quadrant);
if (new Random().Next(0, 100) > 70 && includeRandomization)
{
var randomAction = new Random().Next(0, 100) % possibleActions.Count;
return(possibleActions[randomAction]);
}
else
{
var bestAction = possibleActions[0];
foreach (var action in possibleActions)
{
if (bestAction.Reward < action.Reward)
{
bestAction = action;
}
}
return(bestAction);
}
}
public IList <QAction> GetPossibleActions(QQuadrant quadrant) => Actions.Where(a => a.Current.Equals(quadrant)).ToList();
public QAction(double reward, QQuadrant current, QQuadrant target)
{
Reward = reward;
Current = current;
Target = target;
}
public QEntry(int reward, QQuadrant current, QQuadrant target)
{
Reward = reward;
Current = current;
Target = target;
}