This is the key contribution of deep learning. Developing good representations of state is precisely what today's machine learning is so good at. I agreed with everything you said until here. I think a big component is not really machine learning but more related to how to represent state at any given time, which will necessarily involve a lot of human-tweaking of distilling down what really are the important things that influence winning. I do think that eventually we will get an AI that can beat humans, but it will be a non-trivial problem to solve, and it may take some time to get there. Starcraft is much more open-ended, has many more rules, and as a result its much harder to build a representation of game state that is conducive to effective deep learning. Not to mention that the game itself is more complex, in the sense that go, despite being a very hard game for humans to master, is composed of a few very simple and well defined rules. This alone makes it a much harder problem than go. Starcraft is a continuous game and the game state is not 100% known at any given time. Go is a discrete game where the game state is 100% known at all times. There are several incorrect comments saying that in SC1 AIs have already been able to beat professionals - right now they are nowhere near that level. A lot of people here seem to be underestimating the difficulty of this problem.
0 kommentar(er)
