DeepMind and Blizzard Open StarCraft II as an AI Research Environment

Awhile. This just isn't like Go or Chess. The gap from perfect information to imperfect information is quite a chasm, and from turn-based to real-time is even more vast.

I play Age of Empires 2 semi-competitively, and I just can't imagine the research progress that would have to be made for a pro to lose to an APM-limited AI agent. So much of the game comes down to intuiting what your opponent is planning without being able to see what they're doing, and more importantly intuiting what your opponent isn't ready for.

The biggest difference, though, is the "RT" in "RTS"-- real time. This isn't turn-based anymore, where at a given moment you have a single choice to make, a single piece to move as in Chess and Go, and can then wait for the singular and visible reaction your opponent makes before making your next choice.

My understanding it that the moves a program like AlphaGo makes are not interconnected-- it picks each move individually as an ideal move for that board state. It could take over halfway through the game for someone else and would make the same move that it would have made at that point if it had been in control the whole time and arrived at that board state on its own.

But that doesn't work in a real-time game, since you and your opponent are now moving simultaneously and the "board" is never static. Your moves must be cohesive and planned and flow continuously without time to ponder, each connected to the last. There is no "one" move for a given state.

Another facet of real-time play is the idea of distraction. It's very important in RTS's to keep your opponent distracted, to disrupt their plans and their focus, by coming from unexpected directions at unexpected times, sometimes concurrently with other operations against them. This can't happen in Chess or Go, where the demands on your focus are far less urgent and two things can't happen at once in a literal sense. Can an AI agent learn to appreciate the power of distraction? Can it learn to intuit what will be most disruptive to a human, and what won't be disruptive at all? How can you teach a computer to learn to be annoying?

I will say, of course, that nobody saw AlphaGo coming. And I hope it's the same with RTS's. That would be so exciting. I would love to see an AI blow us away with previously unthought-of strategies. That would be the coolest thing ever. So I hope it happens. But I'd be astonished. RTS is just such a whole new level of thinking for AIs.

> SC2 got rid of a lot of that

You think so? My impression is that SC2 had a lot more of repetitive tasks you had to do. E.g. wall off the ramp, send a worker scouting, ... and you have to perform certain actions every X seconds (like using chrono boost). A lot or mastering the game is rote learning, and polishing a build order. Another big part is constantly scouting and reacting to what the enemy is doing.

Due to those reasons I found SC2 a bit tedious (it was still fun, just felt more like work than SC1). Granted, this is maybe because I played SC1 more on LAN, and there wasn't all the metagame going on. But I think SC2 really does focus on "grinding" and rote learning to get better, probably this was chosen to make it more "eSports"-like.

If I would get to design a SC2.5 or SC3, I would remove all the rote - the actions you always have to perform - and I would give the player the opportunity to trade off more between macro and micro.

Actually, it would be cool if you could "research" certain AI features in game for a cost. For example, have one upgrade that micros your marines like a pro, or positions your units in sensible formations. Another player could counter this with a "radio jam" ability, that would make your units in an area take bad formations, or be controlled by a very simple AI. And if you are good at micro, you could save the update, or invest it in an update that makes macro simpler. And so on, I think there are a lot of things one could explore there. Maybe opening SC2 to AI exploration can lead to such gameplay innovations.

I am so happy to be European... you guys in America always take everything to the total extreme.

Calling a group of women "you guys" is different from calling a person of unknown gender "guy" because in the former case the gender-neutral usage is the only possible interpretation, whereas in the latter it might either be gender-neutral usage or an assumption of the commenter's gender. Given that this ambiguity is nearly always a distraction from the primary purpose of the conversation (as it is here), rightfully or not, pragmatically it is better to use terms that steer clear of it. See Bryan Garner's Modern English Usage for fuller advice on safely navigating this linguistic and sociocultural terrain.

I believe the 400GB is the total amount for the 65000 different game replays