Sort of, yes! Not "solved" in the academic sense of actually finding the mathematically correct/ideal way to play, but rather using neural nets to learn to play and beat humans.