Oh, yes we can. Because that's how these systems learn. There's no learning without a defined reward function.

On the other hand, we could try to imitate the evolutions reward function. But as far as I know nobody really knows how to do that.