Grandmaster level in StarCraft II using multi-agent reinforcement learning

Re:Grandmaster level in StarCraft II using multi-agent reinf... by 6gvONxR4sf7o on Thursday October 31, 2019 @12:13PM (#21410337)

That pales in comparison to starcraft at a higher resolution, which has even more points to click on screen! And even that is nothing compared to playing horseshoes in the physical world. There are an infinite number of moves at any given moment! And an infinite number of moments! Horseshoes is clearly the most complex game of all.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by gpm on Thursday October 31, 2019 @01:09PM (#21411080)

Plank would like to say something to you about lengths.

Which is why the most complex game of all is actually intergalactic horseshoes.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by 300bps on Thursday October 31, 2019 @12:31PM (#21410592)

Starcraft is an incredibly complex game.

Even basic strategies will win if they’re done faster. APM (actions per minute) is a very significant factor into who is winning. Apparently they limited their AI player to 264 APM but that’s still incredibly high and done with machine level consistency. That’s almost 4.5 actions per second!! I know there are human level players at and probably above that level but that really allows for basic strategies to win out.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by gpm on Thursday October 31, 2019 @01:12PM (#21411132)

> Even basic strategies will win if they’re done faster.

This isn't really true. Basic strategies done faster still lose miserably to humans, we can see this from the long history of SC Broodwar AI tournaments where they have a human play the best AI at the end (and always win).

Faster helps, better strategy helps more, maybe we can't say the AI is doing as well as the top humans at strategy since it's faster, but we can say that it's doing pretty good.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by fwip on Thursday October 31, 2019 @03:29PM (#21412672)

AI that ships with games (even the hardest difficulty) are made to be beatable. Of course pros beat them.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by gpm on Thursday October 31, 2019 @04:49PM (#21413584)

You misunderstand me, I'm not talking about the AI that shipped with broodwar. I'm talking about third party AIs made via bwapi competing against eachother in the various starcraft AI tournaments, such as

    AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE)
    IEEE Conference on Computational Intelligence and Games (CIG)
    Student StarCraft AI (SSCAI) Tournament
    BWAPI Bots Ladder

https://bwapi.github.io/

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by fwip on Friday November 01, 2019 @01:38PM (#21421580)

Ah, thank you for the clarification, I did indeed misunderstand.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by tryptophan on Thursday October 31, 2019 @12:58PM (#21410934)

And the bot also can parse the entire screen in .03 seconds and then jump to a new area of the map. No human can monitor the entire map like the bot can.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by baddox on Friday November 01, 2019 @12:04AM (#21416192)

If they want to model the constraints of humans, they probably need to create some “attention” system where the AI needs to choose where to invest its finite attention resources. For example, the AI could choose to focus more on the minimap, which would give them higher reaction times to move around the map, but would reduce their reaction times for things on the main screen.

A great example would be seeing the faint image of cloaked units. How does that work with the AI? Can the just phrase the current screen and instantly see any cloaked unit? I can imagine an attention system where more attention dedicated to part of the screen would increase the probability of noticing a cloaked unit there.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by empath75 on Thursday October 31, 2019 @08:24PM (#21415172)

Sometimes these critiques seem to be complaining that the ai is too intelligent. That’s sort of the point isn’t it.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by bkemmer on Thursday October 31, 2019 @02:40PM (#21412187)

But when the machines start competing with humans it should be fair?

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by frei on Thursday October 31, 2019 @01:48PM (#21411595)

>264 APM

Actually 792 APM. They limited their AI to 264 "things" (choose unit+ability+target or change view) per minute. Each "thing" can be counted as between 0-3 actions by starcraft. So it really has a peak APM limit of 792.

From the article:

>Agents were capped at a max of 22 agent actions per 5 seconds, where one agent action corresponds to a selection, an ability and a target unit or point, which counts as up to 3 actions towards the in-game APM counter. Moving the camera also counts as an agent action, despite not being counted towards APM.

But from replays it has an average APM of 248 at the highest, so nothing crazy going on. Plenty of pros have peak APMs during fights that are higher than 800.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by pps on Thursday October 31, 2019 @01:03PM (#21410994)

264 is not that high. People (if Serral is not a cyborg) can achieve even 1000 APM; and ~300 EPM average for a game.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by SketchySeaBeast on Thursday October 31, 2019 @01:34PM (#21411421)

Those aren't meaningful actions, though, are they? That's 17 actions a second.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by heywot on Thursday October 31, 2019 @01:56PM (#21411701)

Yes and no. There are many repeat actions in SC but at peak they can definitely reach that high. Microing units while simultaneously performing larger macro actions at the same time is a staple of any decent player and is really obvious at the higher levels. I personally play on an average of 100-120 APM for the game but regularly see 200+ in Platinum/Diamond.

[ Parent | on HN ]

Re:Grandmaster level in StarCraft II using multi-agent reinf... by nuclx on Thursday October 31, 2019 @11:25AM (#21409793)

Yes it is, but you can prune the decision space significantly by discarding bogus choices, or more concretely by modelling higher level goals and then compiling them into actual moves.

[ Parent | on HN ]