//function for retrieving the best response for a given initial action
public RPSAction GetActionForStrategy(RPSAction strategy)
{
RPSAction action;
//implement basic rules of the game
switch (strategy)
{
default:
action = RPSAction.Scissors;
break;
case RPSAction.Paper:
action = RPSAction.Scissors;
break;
case RPSAction.Rock:
action = RPSAction.Paper;
break;
case RPSAction.Scissors:
action = RPSAction.Rock;
break;
}
return(action);
}
//handle the overall actions of the game and inform the algorithm of the
//actions from the opponent, in this case, the player's actions
public void TellOpponentAction(RPSAction action)
{
totalActions++;
float utility = GetUtility(lastAction, action);
score[(int)lastAction] += utility;
count[(int)lastAction] += 1;
}
public void TellOpponentActionRM(RPSAction action)
{
lastOpponentAction = action;
for (int i = 0; i < numActions; i++)
{
regret[i] += GetUtility((RPSAction)lastActionRM[i], (RPSAction)action);
regret[i] -= GetUtility((RPSAction)lastAction, (RPSAction)action);
}
}
//computes the next action to be taken
public RPSAction GetNextActionRM()
{
float sum = 0f;
float prob = 0f;
InitRegretMatching();
//explore all available options and hold the response to be taken
for (int i = 0; i < numActions; i++)
{
lastActionRM[i] = GetActionForStrategy((RPSAction)i);
}
//sum the overall regret
for (int i = 0; i < numActions; i++)
{
if (regret[i] > 0f)
{
sum += regret[i];
}
}
//return a random action if the sum is less than or equal to 0
if (sum <= 0f)
{
lastAction = (RPSAction)Random.Range(0, numActions);
return(lastAction);
}
//explore the set of actions and sum the chance of regretting them
for (int i = 0; i < numActions; i++)
{
chance[i] = 0f;
if (regret[i] > 0f)
{
chance[i] = regret[i];
}
if (i > 0)
{
chance[i] += chance[i - 1];
}
}
//computes a random probability and compare that to the chance of taking
//the actions. Returns the first one to be greater than the probability computed
prob = Random.value;
for (int i = 0; i < numActions; i++)
{
if (prob < chance[i])
{
lastStrategy = i;
lastAction = lastActionRM[i];
return(lastAction);
}
}
return((RPSAction)(numActions - 1));
}
public void TellOpponentAction(RPSAction action)
{
totalActions++;
float utility;
utility = GetUtility(lastAction, action);
score[(int)lastAction] += utility;
count[(int)lastAction] += 1;
Debug.Log("Score: Piedra " + score[0] + " Papel " + score[1] + " Tijera " + score[2] + " Lagarto " + score[3] + " Spock " + score[4]);
Debug.Log("Count: Piedra " + count[0] + " Papel " + count[1] + " Tijera " + count[2] + " Lagarto " + count[3] + " Spock " + count[4]);
}
public RPSAction GetNextActionRM()
{
InitRegretMatching();
int i;
for (i = 0; i < numActions; i++)
{
lastActionRM[i] = GetActionForStrategy((RPSAction)i);
}
float sum = 0f;
for (i = 0; i < numActions; i++)
{
if (regret[i] > 0f)
{
sum += regret[i];
}
}
if (sum <= 0f)
{
lastAction = (RPSAction)Random.Range(0, numActions);
return(lastAction);
}
for (i = 0; i < numActions; i++)
{
chance[i] = 0f;
if (regret[i] > 0f)
{
chance[i] = regret[i];
}
if (i > 0)
{
chance[i] += chance[i - 1];
}
}
float prob = Random.value;
for (i = 0; i < numActions; i++)
{
if (prob < chance[i])
{
lastStrategy = i;
lastAction = lastActionRM[i];
return(lastAction);
}
}
return((RPSAction)(numActions - 1));
}
public RPSAction GetNextActionUCB1()
{
int i, best;
float bestScore;
float tempScore;
InitUCB1();
for (i = 0; i < numActions; i++)
{
if (count[i] == 0)
{
lastStrategy = i;
Debug.Log(lastStrategy);
lastAction = GetActionForStrategy((RPSAction)i);
return(lastAction);
}
}
best = 0;
bestScore = score[best] / (float)count[best];
float input = Mathf.Log(totalActions / (float)count[best]);
input *= 2f;
bestScore += Mathf.Sqrt(input);
for (i = 0; i < numActions; i++)
{
tempScore = score[i] / (float)count[i];
input = Mathf.Log(totalActions / (float)count[best]);
input *= 2f;
tempScore = Mathf.Sqrt(input);
if (tempScore > bestScore)
{
best = i;
bestScore = tempScore;
}
}
lastStrategy = best;
lastAction = GetActionForStrategy((RPSAction)best);
return(lastAction);
}
//compute the next action to be taken by the agent
public RPSAction GetNextActionUCB1()
{
int best;
float bestScore;
float tempScore;
InitUCB1();
//check the number of actions available. If an action hasn't been
//explored, return it
for (int i = 0; i < numActions; i++)
{
if (count[i] == 0)
{
lastStrategy = i;
lastAction = GetActionForStrategy((RPSAction)i);
return(lastAction);
}
}
//variables for computing the best score
best = 0;
bestScore = score[best] / count[best];
float input = Mathf.Log(totalActions / count[best]);
input *= 2f;
bestScore += Mathf.Sqrt(input);
//check all actions available
for (int i = 0; i < numActions; i++)
{
//compute the best score
tempScore = score[i] / count[i];
input = Mathf.Log(totalActions / count[best]);
input *= 2f;
tempScore = Mathf.Sqrt(input);
if (tempScore > bestScore)
{
best = i;
bestScore = tempScore;
}
}
//return best strategy
lastStrategy = best;
lastAction = GetActionForStrategy((RPSAction)best);
return(lastAction);
}
//function for computing the utility of an action, based on the opponent's one
//initially it is a draw
public float GetUtility(RPSAction myAction, RPSAction opponentsAction)
{
float utility = 0f;
//check whether the opponent played paper
if (opponentsAction == RPSAction.Paper)
{
if (myAction == RPSAction.Rock)
{
utility = -1f;
}
else if (myAction == RPSAction.Scissors)
{
utility = 1f;
}
}
else if (opponentsAction == RPSAction.Rock)
{
//check whether the opponent played rock
if (myAction == RPSAction.Paper)
{
utility = 1f;
}
else if (myAction == RPSAction.Scissors)
{
utility = -1f;
}
}
else
{
//check whether the opponent player scissors
if (myAction == RPSAction.Rock)
{
utility = -1f;
}
else if (myAction == RPSAction.Paper)
{
utility = 1f;
}
}
return(utility);
}
public RPSAction GetActionForStrategy(RPSAction strategy)
{
RPSAction action;
switch (strategy)
{
default:
case RPSAction.Paper:
action = RPSAction.Scissors;
break;
case RPSAction.Rock:
action = RPSAction.Paper;
break;
case RPSAction.Scissors:
action = RPSAction.Rock;
break;
}
return(action);
}
public RPSAction GetActionForStrategy(RPSAction strategy)
{
RPSAction action;
switch (strategy)
{
default:
case RPSAction.Paper:
action = RPSAction.Spok;
ImagenIA.GetComponent <Image>().sprite = SPOCK;
TellOpponentAction(action);
break;
case RPSAction.Rock:
action = RPSAction.Paper;
ImagenIA.GetComponent <Image>().sprite = PAPER;
TellOpponentAction(action);
break;
case RPSAction.Scissors:
action = RPSAction.Rock;
ImagenIA.GetComponent <Image>().sprite = ROCK;
TellOpponentAction(action);
break;
case RPSAction.Lizzard:
action = RPSAction.Scissors;
ImagenIA.GetComponent <Image>().sprite = SCISSORS;
TellOpponentAction(action);
break;
case RPSAction.Spok:
action = RPSAction.Lizzard;
ImagenIA.GetComponent <Image>().sprite = LIZARD;
TellOpponentAction(action);
break;
}
return(action);
}
public float GetUtility(RPSAction myAction, RPSAction opponents)
{
float utility = 0f;
if (opponents == RPSAction.Paper)
{
if (myAction == RPSAction.Rock)
{
utility = -1f;
}
else if (myAction == RPSAction.Scissors)
{
utility = 1f;
}
}
else if (opponents == RPSAction.Rock)
{
if (myAction == RPSAction.Paper)
{
utility = 1f;
}
else if (myAction == RPSAction.Scissors)
{
utility = -1f;
}
}
else
{
if (myAction == RPSAction.Rock)
{
utility = -1f;
}
else if (myAction == RPSAction.Paper)
{
utility = 1f;
}
}
return(utility);
}
