public override double Playout(int horizon)
{
// refrac: do not believe rewards that are further away so much
double totalReward = 0.0;
for (int i = 0; i < horizon; i++) {
int action = this.GenerateRandomAction();
this.ModelUpdateAction(action);
var or = this.GeneratePerceptAndUpdate();
totalReward += or.Item2;
}
return totalReward;
}