AlphaStar: Mastering the Real-Time Strategy Game StarCraft II • Hacker News Board

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#241

Post by a_wild_dandan » Fri, Jan 25, 2019, 2:28 AM UTC

gamegoblin wrote: post #67

rcheu wrote: post #46

This is really impressive, I didn't expect starcraft to be played this well by a machine learning based AI. I'm excited to read the paper when it comes out! That said, I'm not sure I agree that it was winning mainly due to better decision making. For context, I've been ranked in the top 0.1% of players and beaten pros in Starcraft 2, and also work as a machine learning engineer. The stalker micro in particular looked…

In the mass stalker battles, the AI APM exceeded 1000 a few times, and no doubt that most of that was precisely targeted. Whereas a human doing 500 APM micro is obviously going to be far more imprecise. I think a far more interesting limitation would be to cap APM at 150 or so, or to artificially limit action precision with some sort of virtual mouse that reduced accuracy as APM increased.

Here's a graph of AlphaStar's APM versus a professional player's: https://i.imgur.com/TXeLkQK.png Evidently AlphaStar also has a similar Economy of Attention (where the player focuses) to a professional player, at around 30 screens per minute. Additionally, AlphaStar's reaction time is around 350ms, a significant disadvantage over a pro.

The skepticism in this thread is absolutely justified but I think it's important to note the lengths to which DeepMind has gone to address and assuage the fears of superhuman mechanical skills being employed in these games.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#242

Post by Ar-Curunir » Fri, Jan 25, 2019, 2:29 AM UTC

derefr wrote: post #223

asteli wrote: post #205

Earlier quoted context omitted.

Perhaps a nit, but still fascinating: the human visual cortex finishes developing after birth. A newborn can't really distinguish between objects. The ability to differentiate, focus on and track objects is developed over the course of several months.

True. Humans are pretty unique in that regard, though; pretty much no other animal is like that. It's easier to understand human neonatal development if you just considering all humans to be born premature. (It'd be really interesting to know whether that's literally true—whether keeping a human baby in the womb for an extra few months would actually result in the same stages of mental development being passed that o…

I've read somewhere that we are basically born prematurely (as you said) because if we waited any longer then our enlarged head sizes would make delivery quite possibly fatal.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#243

Post by owens99 » Fri, Jan 25, 2019, 2:31 AM UTC

Interested to hear what others here think.

The APM metric includes all clicks a player makes. AlphaStar's APM is lower than a typical pro's, but that does not mean it is making fewer actions. All pros try to keep a high "tempo" by constantly clicking the screen even when they are not making any actions. e.g. Instead of sending a unit to a specific location with one click, they will click 5-6 times while dragging the mouse to that spot. The theory is that keeping a high tempo allows you to make more useful actions overall.

Unless AlphaStar's average and maximum APM is 3-4x lower than a pros, rather than just 2x lower average and the same maximum, I do NOT believe that this is a fair test of the AI's strategic decision making ability.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#244

Post by setr » Fri, Jan 25, 2019, 2:31 AM UTC

CydeWeys wrote: post #177

arcticfox wrote: post #166

Earlier quoted context omitted.

> How many of those 500 actions are actually useful? Exactly, a human doing 500 APM during intense moments is going to be way different than an AI bursting 1000 APM with pixel-precision during the most crucial moment in a game. TLO spent a ton of time at >1000 APM and walked his army directly into enemy shots all the time. MaNa had much better control at ~400 APM. So APM is really irrelevant to control - for humans.…

I suspect that once the AI becomes good enough it will be able to beat human players using a much lower total APM than human players. We're not quite there yet, but it just needs a little bit of time. As a hopefully illustrative comparison, you could give any top player a day of play time per move against the top Chess AI being given a minute of play time per move and the AI will still win. That's how much better the…

The biggest issue with allowing the ai to have high APM is that it will inevitably learn optimal strategies that depend on that high APM, eg stalkers can take on far more immortals than we normally expect, and the AI will learn it this way, because the high APM allows a new stalker strategy (or rather, empowers an old one greatly) while not affecting immortals significantly. This also naturally means the AI leagues see a different game balance than the human leagues, leading to strategy divergence.

And then when you drop the APM limit, suddenly all the learned optimal ai strategies start falling apart, and the whole thing has to be relearned.

More annoyingly, there’s not much for human players to learn from innovative ai strategies that are based on inhuman accuracy of play (because we couldn’t possibly execute it).

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#245

Post by arnaudsm » Fri, Jan 25, 2019, 2:32 AM UTC

Today, 1/1 games without the camera hack were won by the human.

This is disappointing, but their PR is a success. Mainstream media is already saying "AI has beaten StarCraft". Another lie for the AI bubble I guess.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#246

Post by foobiekr » Fri, Jan 25, 2019, 2:33 AM UTC

methodover wrote: post #121

olliej wrote: post #78

Earlier quoted context omitted.

it seems like in some cases at least it didn't have to move the camera (it had direct interfaces) which for some of the stalker micro battles (especially in game 3 or 4?) the battles were larger than the screen space -- it would not have been possible to micro that well if your control interface limited what you can control or where you can place them.

This is a great point, and something that seems a bit lost in the discussion: In StarCraft 2, the game IS the interface. That is to say, the developers have constructed the game in such a way as to be difficult to control; and human mastery of the interface is a large percentage of the game. Strategy in the game is important, of course -- but this is not chess, where human beings are not limited by the interface of t…

Are there any bolt-on augmentation interfaces that utilize the same API the bots use to allow players to more effectively enter their intent?

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#247

Post by spydum » Fri, Jan 25, 2019, 2:34 AM UTC

dx87 wrote: post #181

jgon wrote: post #150

Earlier quoted context omitted.

Even the 200ms reaction time seemed overly slanted towards the AI. I don't think that is the actual reaction time of top pros, in the matches the AI played the human player would teleport in from complete invisibility and try to use an instant cast spell and the AI would have already teleported out. Yes the theoretically may have been constrained to a 200ms reaction time, but in practice the AI was playing at a super…

Another advantage was that the AI is just reading the game state through an API, it doesn't have to look on the screen. The game can be difficult to watch from a pro's perspective since they have to constantly click around the map to see what's happening, but the AI has perfect knowledge of everything it is capable of seeing, all without having to physically move a mouse to click on the screen.

If you watch the 11th game where pro player wins, (prior was a 10-0 shutout by Alpha), the AI actually lost because they rebuilt the agent to use the same forced camera perspective as the human - so there is absolute truth to this being a compelling advantage. It was able to micro multiple units in disparate areas by having far better spatial awareness. When they took that advantage away it seemed more even.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#248

Post by cjbprime » Fri, Jan 25, 2019, 2:37 AM UTC

derefr wrote: post #182

dcl wrote: post #164

Earlier quoted context omitted.

This is a widely underappreciated fact when it comes to comes to comparing the 'training experience' of humans versus bots. And it extends far beyond processing 'sense data' - A human likely has some level of understanding of how the game works based on experience from other games it has played and from 'real life' - we know almost instinctively that 'high ground' is likely to give a combat advantage without having t…

Not only that, humans (and many other eusocial species) have an instinctual intuitional understanding of many aspects of game theory. For example, humans, even from infancy, prefer games where it is possible to punish cheating (i.e. take revenge upon cheaters) to games where it is not. This isn't just "we're animals that have evolved to enact tit-for-tat strategies [by e.g. injustice triggering rage] because they lea…

And perhaps worth noting that the great apes we evolved alongside have the same kind of outrage to unfair trades.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#249

Post by cjbprime » Fri, Jan 25, 2019, 2:39 AM UTC

vertline3 wrote: post #191

arcticfox wrote: post #166

Earlier quoted context omitted.

> How many of those 500 actions are actually useful? Exactly, a human doing 500 APM during intense moments is going to be way different than an AI bursting 1000 APM with pixel-precision during the most crucial moment in a game. TLO spent a ton of time at >1000 APM and walked his army directly into enemy shots all the time. MaNa had much better control at ~400 APM. So APM is really irrelevant to control - for humans.…

Such high actions per minute does not seem fun to me, and possibly a repetitive strain injury waiting to happen.

Wait until you hear about stringed musical instruments? :)

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#250

Post by celeritascelery » Fri, Jan 25, 2019, 2:39 AM UTC

Here is an example of what you do without artificial limitations on actions. https://youtu.be/3PLplRDSgpo Just goes to show how important those restrictions are for competitive gameplay against AI.