protected override Reward Playout(WorldModel initialPlayoutState)
{
/*
* while s is nonterminal do
* chose a from Actions(s) uniformly at random
* s <- Result(s,a)
* return reward for state s
*/
FutureStateWorldModel currentState = initialPlayoutState.GenerateChildWorldModel() as FutureStateWorldModel;
GOB.Action randomAction;
ActionHistory.Clear();
int currentDepth = 0;
Pair <int, Action> par = new Pair <int, Action>(0, new Action("asdasdas"));//cria com lixo para depois ser substituido
while (!currentState.IsTerminal())
{
randomAction = currentState.getNextBiasRandomAction(this.RandomGenerator, currentState);
randomAction.ApplyActionEffects(currentState);
currentState.CalculateNextPlayer();
par.Left = currentState.GetNextPlayer();
par.Right = randomAction;
ActionHistory.Add(par);
currentDepth++;
}
if (currentDepth > this.MaxPlayoutDepthReached)
{
this.MaxPlayoutDepthReached = currentDepth;
}
return(new Reward()
{
PlayerID = currentState.GetNextPlayer(), Value = currentState.GetScore()
});
}