User:OmaMorkie/The AI was beating us in Go. And then we changed the Meta.

The AI is coming for Starcraft, but we have some early scouting on it

In general terms, what the AI's in Go and Chess are really bad at is Meta. I understand just about enough go to see that in the commentary on the pro-go games with and against AI. The AI invented a new move that killed the current meta. I think it was number 38 or something in the first game, where the human took a 2 hour break before making his move – that's like half of the entire game's thinking time. The moved altered something in the meta of Go. But the humans where faster to start adapting the Meta, and that is the only way then can at least occasionally win a game. Changing number 38 is a bit like adding 6 burrowed Banelings in the mineral line of your opponents 3rd before he takes it. A really small detail in the bigger picture. But it may be enough to break the balance in favor of Zerg when used in an otherwise perfectly executed hydra-ling-bane in ZvP. Once that piece is broken, all the logic about the early-game strategy changes leading to that point also change. And the next step after that is when humans prevail: Adapting the Meta. If I really understand enough Go -lingo, I believe that this is exactly where the human wins lie: They change up something in the early game to a new move that was considered impossible long ago. But they somehow make it work. This makes sense when looking at the technical way the AIs the game at the moment: They need a very, very large library of games to get going. We are still ahead when it comes to creating a small number of viable new gamestyles in that library as soon a new change pops up. That is because we have a good theory of how the game works, while the AI is only experimenting based on re-enforcement learning. Or can they develop theories? Or experiment so fast that they don't need one? I don't even know what that would mean… One way or another:

So let's stab at the one know weakness of the AI's: Speed up on improving the Meta.

Let me briefly define what I mean with the “Meta”. The terms is generally used to describe the “expected” reactions of your opponent. For example, you can say: “Battlecruisers are currently not in the meta in TvT”, which means you won't see battlecruisers in like 95% of Grand-Master game winners. The meta is composed of a set of build-orders for each race, which are all connected with a set of “reasonable” responses. A “Build order” is the literal order of buildings, but let's include certain non-building “meta-moves” like purposefully engaging in a non-efficient trade in the definition to keep it more broad.

How can we do that?

That is easier said than done… Aren't we all trying all the time to get ahead in the Meta? I'd even claim that that is what makes Starcraft so great and let's us stick with the same fucking game for decades...? We are doing it, yes, and we are even quite advanced at it, so we have a language to talk about, it's just not very precise. While I'm sure there is a lot of talk about exactly the kind of weird move like 6 burrowed banelings in your opponents third mineral line before he takes it, we do not take records or systematize how we understand Starcraft. I mean, we don't have a real “Theory of Starcraft” that lists all the relevant factors (like the theories used to creating scoring scheme of a situation in Go or Chess). Or rather: There is a pretty advanced theory out there that we somehow share so we can talk about these things and know a lot, but we never formalized that. The AI's will learn all that about the Meta that know from previous games and start cornering us due to their better execution, and the one thing were our advantage is biggest is developing the meta.

Developing the Science of Starcraft

One way to formalize that talk would be to develop a code that allows us to talk faster. We have an informal code that is really sophisticated already – like in when I make a statement like: “Can you believe it? Stats just won the GSL after seven minute final against Gumiho when he did the Void Ray & Basetrade move that Strange had used in that “Strange vs. Polt” game”. Most Starcraft casters and many followers will have understood right away which move in which game I'm referring to, even if it's already more than a year old.

Let's start with the language used to describe Starcraft.

I would argue that what Strange did in Strange vs. Polt was a potentially Meta altering event. There are many others, so I'm trying to formalize what he said here. A more frequently occurring one would be “Neebs Archon Drop”, but I'll stick to Strange vs. Polt as an example for now. The more subtly we can improve the Meta, the better. Defending Archon Drops with Roaches at perfect execution is something so clearly possible that the AI would find that by just experimenting a bit beyond a fully standardized, perfect execution ZvP usual game.

As common in science 101 books, I guess we need to define a way to talk efficiently first. Talking efficiently is one way to improve our understanding, in the same way a scientific theory is better when it allows more precise descriptions of events. I'm thinking of saying the “what strange did in polt vs. strange” sentence from the last paragraph in less words basically, and without creating issues like if there could ever be another crazy memorable polt vs. strange game. I'm really not sure how to do that exactly, maybe something like: “(red; blue; date; gametime; other factors)” could be nice, like “(Strange; Polt; 15:00; DH16; RO32)” referring to that game. But we also need a more efficient way to talk about why that game mattered. Like build-orders and game statistics. Those are super formal and detailed, so maybe great for analysis, but not for talk. But we've done pretty good so far and also have the way we normally talk about games already. Saying “Stats just won the GSL against Gumhio by initiating a base trade against Terran with Void Rays at 6 minutes” works pretty good. Probably that game would also end up the Hall of Fame of crazy games, so that would make it “Stats won in (Stats;Gumiho; 6:00;BC17; GF), because he did…what? Maybe “BaseTrade in PvT with Void-Rays at 6 minutes” – codified that would make it “Strategy in MatchUp in TechDeviation from Meta”. Would be very curious to see high-level commenting and explaining of Meta-Chaning games by our top-youtubers (like PiG or Lokow) translated into such a more formal language.

But, more importantly start experimenting

But as all sciences, Starcraft will for now be an experimental driven science. We have no way of telling if (BT; PvT;VoidR; 600) is opening up a new avenue for P to get a larger % in wins or not. If it is a strategy that will actually take off to professional level.

We need more games like (Strange; Polt; 15:00; DH16; RO32). And you know how probably would like to see more games like that? The casters and the viewers. I didn't check the numbers, but StrangeVSPolt pretty much was the most seen cast like ever. But games like that are rare, and move like Strange's even rarer. Our pro-gamers cannot rely on the scientific axiom of “perfect execution”, so they can't just do something they never tried under high pressure to win. We already see a lot more experimentation among the Pro's non-price games. But could there be a way to push our pro's to focus less on execution and more on Meta? They will have too, as soon as the Robo-Hybrids enter the ladder that skill looses value. But money makes the world go round, so let me just propose the first Starcraft research policy:

A crowd-sourced award for the coolest game.

As the crowd apparently really, really liked Polt vs. Strange, and the casters and you-tubers even made actually more money because of that particular game, we could launch a second reward to the biggest tournaments. One that is rewards the player who made the most unorthodox move work. You can vote which move that was based on how much money you put in & with everybody how chips in have all the replays.… That, and the crowd, should be all it takes to get one, or maybe two, very important things in Starcraft II:

<p lang="en-US" style="margin-bottom:0cm;line-height:100%">1) Better chances against the robots

<p lang="en-US" style="margin-bottom:0cm;line-height:100%">2) Epic Games

<p lang="en-US" style="margin-bottom:0cm;line-height:100%">

@page { margin: 2cm } p { margin-bottom: 0.25cm; line-height: 120% } a:link { so-language: zxx }

<p lang="en-US" style="margin-bottom:0cm;line-height:100%">This is it for now from Oma Morkie's Starcraft Science Blog, but there is already my first policy recommendation.