10 ms·

This is pretty interesting. DeepMind’s last triumph (beating the best human Go players with AlphaGo) is impressive, but Go is a great fit for neural networks a

• andrew3726 10y ago
  > When Google’s AI can beat a human at StarCraft, it’s time to be very afraid. More like a time to be very excited! But I totally agree, long-term dependencies are an difficult problem. The DNC[0] by DeepMind is in my opinion a very important step in the correct direction to tackle this. [0] https://deepmind.com/blog/differentiable-neural-computers/ https://deepmind.com/blog/differentiable-neural-computers/
  • chmod775 10y ago
    Honestly you can solve fog of war using regular computing (TM) to remember the last known state of a particular area, then feed that to the network along with information that A. it's an area as it was seen some time ago B. how much time elapsed since it was last uncovered. I can't imagine that this is where DNCs come in unless you want to needlessly complicate the matter (and create a network that is much more expensive to train).
    • thomasahle 10y ago
      Yes, it's interesting what help will be given to the network. From the article it seems the answer is: very little. It is allowed to use only the pixels on the screen. (During play, not training apparently.) It will even have to spend actions on moving the screen/window to inspect things. Very cool stuff :)
• burke 10y ago
  > When Google’s AI can beat a human at StarCraft, it’s time to be very afraid. I don't actually agree with this. Unlike Go, StarCraft is not only a game of strategy; "micro" (micro-managed tactics, basically) also plays a big role. An AI is going to be able to issue a LOT more commands per second than even the most skilled humans, giving them a natural tactical advantage. Strategy is more difficult for an AI, but not the only deciding factor in StarCraft.
  • arcticfox 10y ago
    I agree. I think you'd have to limit that machine's APM to somewhere around the human limit, and make it choose which actions it wants to spend its time on. Otherwise the AI is eventually just going to absolutely destroy humans with micro, which is not remotely interesting. We've known transistors were faster than fingers for ages.
    • lordnacho 10y ago
      > Otherwise the AI is eventually just going to absolutely destroy humans with micro, which is not remotely interesting. If you could issue as many orders as you wanted, what would the optimal strategy be?
      • Thaxll 10y ago
        Simple example would be in combat, like being always out of range ect ... Imagine one army where every single unit is controlled by one human. Each individual unit would be very very strong.
        • Maro 10y ago
          > Imagine one army where every single unit is controlled by one human. Each individual unit would be very very strong. Like a real army :)
      • trjordan 10y ago
        A big part of Starcraft is that certain units counter other units. Tanks counter zerglings, because they do splash damage to a bunch of them at once. Here's what you can do with infinite APM: https://www.youtube.com/watch?v=IKVFZ28ybQs https://www.youtube.com/watch?v=IKVFZ28ybQs It's just broken. It's not even the same game.
• wingcommander 10y ago
  There is so much sensationalist nonsense by clueless people here, it is insane. Do you freak out over everything you do not understand? Because you don't understand the solid state physics of your iPhone CPU, does that mean it is magic? Because you do not understand what a neural network is, does that mean it is magic? NO, it is essentially curve fitting with gradient descent. (Me: PhD in Statistics).
• kevinwang 10y ago
  your last sentence, and this comment in general sounds exactly like what people were saying a year ago about Go. (Chess was easy, because it's perfectly suited for AI, but when a computer can beat us at Go, that's when we should be afraid!)
• empath75 10y ago
  Even if you don't imagine robots becoming self aware and deciding to kill humans on their own, you can imagine a fleet of drones and self-driving tanks controlled by a deep learning network. They don't need to understand what they're doing to make effective tactical decisions. If the US doesn't do this, someone else probably will. We're going to see an AI-based blitzkrieg attack in our lifetimes. Once you can launch an invasion without losing any soldiers, war seems like a much less dangerous proposition.
  • qznc 10y ago
    We have not seen the bomb-carrying drones yet. That is entirely possible right now, but IS has not used that so far.
    • theptip 10y ago
      Actually, this has happened already: https://www.washingtonpost.com/news/checkpoint/wp/2016/10/11/isis-used-an-armed-drone-to-kill-two-kurdish-fighters-and-wound-french-troops-report-says/ https://www.washingtonpost.com/news/checkpoint/wp/2016/10/11...
    • empath75 10y ago
      The us has been using them for decades.
  • visarga 10y ago
    > Once you can launch an invasion without losing any soldiers, war seems like a much less dangerous proposition. Yes, but how can you stop everyone else from doing the same, if the AI part will be trivial to replicate?
    • taneq 10y ago
      Once one group has an AI capable of managing a drone army in near-optimal fashion, there won't be an 'everyone else'.
      • 10y ago
• mmanfrin 10y ago
  Really rudimentary AIs are already significantly beyond the skill of humans (in terms of micro, at least). https://www.youtube.com/watch?v=IKVFZ28ybQs https://www.youtube.com/watch?v=IKVFZ28ybQs
  • dwaltrip 10y ago
    They are quite terrible at all of the other aspects: macro, broad strategy, tempo & timing, planning tactical surprises, etc. This results in a very low overall competency, when it comes to winning games against highly skilled opponents. It will be a challenge for researchers to improve performance in these areas. The problem space is very complex -- it makes solving Go look like a cake walk in comparison.
  • ygra 10y ago
    As you note, that's only micro, and it's the easy part since you can do most of that with rather simple rules and just high apm (which is where the benefit compared to human players comes from). The interesting part for AI is how to get there, the longer-term planning, recognizing the enemy's strategy based on what it observes, reacting appropriately, etc. So far those are the things AIs have trouble with.
• Plough_Jogger 10y ago
  There is a growing body of research surrounding deep reinforcement learning that would reject the idea that "[ neural networks ] are not a good fit for most real-world problems, where you have to remember something that happened in the past."
• imaginenore 10y ago
  AI already beats humans at StarCraft, not because it's smarter, but because it can control all the units simultaneously. https://www.youtube.com/watch?v=IKVFZ28ybQs https://www.youtube.com/watch?v=IKVFZ28ybQs https://www.youtube.com/watch?v=3PLplRDSgpo https://www.youtube.com/watch?v=3PLplRDSgpo https://www.youtube.com/watch?v=XgfTpU-a8D0 https://www.youtube.com/watch?v=XgfTpU-a8D0
• shannonpig 10y ago
  I think you're thinking about the difference between full information games vs partial information games, rather than stateless games (the state of a starcraft game is stored on a central server, the computer just has no access to it). There's already been excellent progress in poker AI based on traditional game theory, see http://poker.cs.ualberta.ca/ http://poker.cs.ualberta.ca/.
  • keerthiko 10y ago
    It's not just the limited information space, however. There's also the solution space or "action space". A poker game-space is quite narrow compared to that of StarCraft because of pixel-wise differences. This is the main reason that makes StarCraft "much more realistic". In real life you have infinite moves and options with (almost) analog differences, whereas a game like Poker is much more discrete. Go is somewhere in between. For instance, as any protoss player knows this all too well, a Zealot placed a mere pixel away from where it should could allow a stream of zerglings in, completely throwing any game plan you had up to that point out the window and enter crisis management mode. Usually this is a "mistake", which a computer may never make, but it needs to learn the relevance of such pixel-sized mistakes, which is unlikely to have as much of an impact in a poker game or Go game as it would in StarCraft.
    • gspetr 10y ago
      > but it needs to learn the relevance of such pixel-sized mistakes, which is unlikely to have as much of an impact in a poker game or Go game as it would in StarCraft. I agree with your point, but you need to look no further than the game that AlphaGo lost against Lee Sedol. It committed a very bizarre mistake that made no sense gamewise.
  • theptip 10y ago
    You're right, that's the technical term for it, thanks. I was referring to the agent's evaluation function being stateless (i.e. it does not _need_ to persist state from previous turns in order to play optimally). I think that there are two distinct aspects of the incomplete information that are significant; first, ignoring temporality, you have the fog of war, so you can't see the whole board. This is probably easier to address in a RNN, since you can play quite well as an amnesiac that just reacts to things that are currently visible. But you need to scout less if you have a memory of what's out there, so the amnesiac won't be able to play optimally. Then there's the temporal aspect. The set of previous states of the game is not stored in the game and made available to the player, and so to play optimally you have to have a memory. This is where new techniques will be necessary. These are separate problems, I think, so it will be interesting to see if DeepMind can make progress without reaching human performance on the second part.
• DonaldFisk 10y ago
  > it’s stateless, so you can fully evaluate your position based on the state of the board at a given turn. That's not always true: https://en.wikipedia.org/wiki/Ko_fight https://en.wikipedia.org/wiki/Ko_fight