On the other hand, the shooting template can be extended to an adversarial situation by putting in a defender. In order to make this task fair, we widen the goal by 3 times (see Figure 4(b)). During our initial experiments, the shooter learned to shoot at the upper and lower portions of the goal as well as the middle. So the shooter can aim away from the defender; the defender must try to figure out where the shooter is aiming and block that portion of the goal.
The defender may be able to learn to judge where the shooter is aiming before the shooter strikes the ball by observing the shooter's approach. Similarly, if the defender starts moving, the shooter may be able to adjust and aim at a different part of the goal. Thus as time goes on the, opponents will need to co-evolve in order to adjust to each other's changing strategies. Note that a sophisticated agent may be able to influence the opponent's behavior by acting consistently in the same way for a period of time, and then drastically changing behaviors so as to fool the opponent.