public float GetUtility(RBYAction lastAction, RBYAction action)
{
float utility = 0;
if (lastAction == RBYAction.Red && action == RBYAction.Red)
{
utility--;
}
else if (lastAction == RBYAction.Red && action == RBYAction.Blue ||
lastAction == RBYAction.Red && action == RBYAction.Yellow)
{
utility++;
}
else if (lastAction == RBYAction.Blue && action == RBYAction.Blue)
{
utility--;
}
else if (lastAction == RBYAction.Blue && action == RBYAction.Red ||
lastAction == RBYAction.Blue && action == RBYAction.Yellow)
{
utility++;
}
else if (lastAction == RBYAction.Yellow && action == RBYAction.Yellow)
{
utility--;
}
else if (lastAction == RBYAction.Yellow && action == RBYAction.Red ||
lastAction == RBYAction.Yellow && action == RBYAction.Blue)
{
utility++;
}
return(utility);
}
private RBYAction GetNextActionUCB1()
{
int i, best;
float bestScore;
float tempScore;
for (i = 0; i < numActions; i++)
{
//spawneo inicial aleatorio
if (count[i] == 0)
{
int r = Random.Range(0, 4);
lastAction = (RBYAction)r;
return(lastAction);
}
}
//aplica UCB1
best = -1;
bestScore = int.MinValue;
for (i = 0; i < numActions; i++)
{
tempScore = UCB1(score[i] / count[i], count[i], totalActions);
UCB1scores[i] = tempScore;
if (tempScore > bestScore)
{
best = i;
bestScore = tempScore;
}
}
lastAction = (RBYAction)best;
return(lastAction);
}
public void TellOponentActionUCB1(RBYAction action)
{
totalActions++;
float utility;
utility = GetUtility(lastAction, action);
score[(int)lastAction] += utility;
count[(int)lastAction]++;
}
public void PlayerPlay(int action)
{
RBYAction iaAction = ia.Play();
if (ia.GetUtility(iaAction, (RBYAction)action) == 1)
{
scoreIA++;
}
else if (ia.GetUtility(iaAction, (RBYAction)action) == -1)
{
scorePlayer++;
}
ia.TellOponentActionUCB1((RBYAction)action);
}