Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

DeepMind and Blizzard Open StarCraft II as an AI Research Environment

deepmind.com

Post reply on HN
Print view View original

91–100 of 283 posts

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#91

Post by vmasto »

pavedwalden wrote: post #88
yannyu wrote: post #85

Earlier quoted context omitted.

Why? Speed is in MPH or KPH, would it be better to go by m/s? The convention in gaming is APM, so they're just using the nomenclature that is already understood.

Maybe what usaphp is getting at is that the AI could still gain an advantage by doing a set of actions much faster than humanly possible in just a fraction of a second as long as it kept its total number of actions that minute below the cap.

Exactly. Most actions in idle game mode (i.e. when there's no active battle, micro or macro to be done) are null, the gamers simply repeat random meaningless keystrokes; this helps with being always alert and ready for anything that requires a rapid response.

An AI doesn't need to do this.

Source: Ex pro Starcraft (Brood War) gamer.

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#92

Post by cjbprime »

infinite8s wrote: post #56
tantalor wrote: post #54

Earlier quoted context omitted.

Watson’s hand looks like a clear, Plexiglas, cylindrical soda can with a few metal screws in the top and a wire extending from the bottom that is connected to Watson’s Front-End Controller. The mechanical hand wraps around the Jeopardy! buzzer which is inserted in the bottom of the Plexiglas cylinder and is held in place by a clamp. Watson’s hand uses a solenoid to physically press the same button that the humans mus…

Typical human response time to a stimulus is about 200ms. That of course doesn't factor in the decision time, but once the brain decides on an action it takes about 200 Ms for the signal to propagate to the relevant muscles.

How about exceptional human response time, as for pro gamers or Jeopardy players who have predicted the stimulus?

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#93

Post by daemonk »

ajkjk wrote: post #32
daemonk wrote: post #6

Blizzard should put in an AI-assisted play mode where players are limited to X lines of code that can be launched with keyboard commands.

I know that, as a player, the high mechanical limitations of Starcraft are part of why it's such a difficult, high-skill-ceiling game. But.. I've tried to enjoy watching SC2 on Twitch, and while it's kinda fun, it's just so disappointing when a complicated strategic game is thrown away because a player doesn't react fast enough to workers being sniped or a drop being shot down. I wish the individual units had some au…

Yeah I can even imagine a thriving "marketplace" for specialty code that top players would keep secret.

And it doesn't have to just be for micro. For people who are bad at macro, maybe code can be written to consistently maintain X workers at all bases.

The difficult part here would be how to balance the AI-assistance. Is lines of code (or number of characters) a good proxy for complexity? What's the number-of-character to benefit ratio?

I guess that's ultimately determined by the individual player's strengths and weaknesses. If a player sucks at macro, then the macro script is worth the number of characters.

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#94

Post by ktRolster »

ehsankia wrote: post #77
EGreg wrote: post #50

Earlier quoted context omitted.

I would like to see IBM Watson dominate Jeopardy without a superhuman finger.

Yes and no. I guess the more "fair" way to compare them at trivia would be to have a format where all 3 give an answer to every question, and whoever gets the most points wins. That's obviously not really jeopardy, and I'd be curious to know how Watson would perform purely on the trivia knowledge vs clicking speed. I'm guessing it would still win though.

"What is your mother's name?"

"......let me tell you about my mother...."

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#95

Post by metachris »

JefeChulo wrote: post #30

"so agents must interact with the game within limits of human dexterity in terms of “Actions Per Minute”." I am really glad they are limiting APM because otherwise things just get stupid.

Which is still about 350 actions per minute for a professional SC2 player, roughly 5 to 6 actions per second.

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#96

Post by ktwo »

JefeChulo wrote: post #30

"so agents must interact with the game within limits of human dexterity in terms of “Actions Per Minute”." I am really glad they are limiting APM because otherwise things just get stupid.

IMO there should also be a precision limit. The timing of actions should include human-typical jitter and the wrong action should sometimes be activated to simulate misclicks/fat-finger keypresses — e.g., messing up a control group by assinging a unit to the wrong number key. The bot must also not be able to act faster than human reaction times (~250ms), this could be enforced by adding a fixed delay to the observations.

I wouldn't be surprised if human Starcraft II play isn't so much limited by decision-making as by the translation of decisions into mechanical actions, which in turn dilutes the attention devoted to actual decision making.

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#97

Post by colmvp »

Synaesthesia wrote: post #81
hitekker wrote: post #59

This seems all in good fun but I wonder if it's come too late. Starcraft 2 is at its twilight. The biggest leagues of South Korea have disbanded. [1] The prolific progamers who transitioned to Starcraft 2 have gone back to Broodwar. [2] Blizzard itself has scrubbed all references to Starcraft 2 on the very home page of Starcraft. [3] Except for the twitter embed, it has only only one "2" character... in the copyright…

I disagree I got into Starcraft recently and find it very much vibrant, both in the pro scene and casual. But that’s irrelevant. The point is it’s still a great ai challenge

It's a great AI challenge but the pro/casual scene is very diminished from what it once was. Practically every single streamer who introduced me to the concept of streaming by playing/casting SC2 have either moved onto other games or quit streaming altogether. I can't believe it was only a few years ago, but I used to watch Husky/HD/Day9 everyday.

Unless I'm mistaken, even the top SC2 streamers today receive a fraction of what other streamers who stream games like Hearthstone or Dota2 get. I'm not suggesting it's a 'ded gaem' but to me it's become a little like AoE2 in that it's a niche e-sport, which is certainly nothing to be ashamed about. But I think it's a far from what Blizzard had hoped for, which I think is reflected with their next batch of games in Hearthstone, Overwatch, and HotS, which all have some level of competitive play while still being way more very friendly to casual users than SC2.

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#98

Post by solicode »

hitekker wrote: post #59

This seems all in good fun but I wonder if it's come too late. Starcraft 2 is at its twilight. The biggest leagues of South Korea have disbanded. [1] The prolific progamers who transitioned to Starcraft 2 have gone back to Broodwar. [2] Blizzard itself has scrubbed all references to Starcraft 2 on the very home page of Starcraft. [3] Except for the twitter embed, it has only only one "2" character... in the copyright…

Personally, I think focusing on BW would have been more interesting (as long as the APM limit still stands), but I guess SC2 is alright too. The fact that they're even doing this though makes me happy.

The reason I say BW would be especially interesting is simply because the game has remained basically unchanged balance-wise since v1.08 which came out in 2001. Despite that, the pro scene never left, and we're still seeing some shifts in the meta even today. It would be cool to see a strong AI flip the script completely for such an established and "well understood" game. Opportunities like that are kind of rare, at least when it comes to video games.

Re: DeepMind and Blizzard Open StarCraft II as an AI Research Environment

#99

Post by arcticfox »

artursapek wrote: post #68
aerovistae wrote: post #41

Earlier quoted context omitted.

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 bei…

Speaking of AoE 2, I really hope this research ends up benefiting that game. The AI in it has always been so bad but it's still my favorite game of all time.

Microsoft hired one of the best scripters (Promi) on the AoE2 AI circuit to write the official AI for AoE2 HD, and it's quite good, although the rules engine doesn't have enough features to beat early harassment. If you let it boom it's pretty scary for new players.

If you play using the UserPatch on Voobly, where the serious custom AIs are written, you can play vs. Barbarian. It is _very_ good, if you're not a semi-competitive player it will certainly beat you.

The upcoming UserPatch 1.5 will add even more features, so Barbarian and other custom AIs will become stronger.

To be clear, AoE2 AIs are all rules-based and written by pro players themselves, which is quite different from what DeepMind is trying to do.

Post reply on HN