Slashdot Mirror
StarCraft AI Competition Announced
bgweber writes "The 2010 conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE 2010) will be hosting a StarCraft AI competition as part of the conference program. This competition enables academic researchers to evaluate their AI systems in a robust, commercial RTS environment. The competition will be held in the weeks leading up to the conference. The final matches will be held live at the conference with commentary. Exhibition matches will also be held between skilled human players and the top-performing bots."
Or just smarter than the random BNET player?
Let's teach our AI systems how to do battle... against humans. Skynet anyone?
So they are using the old StarCraft and not the new upcoming StarCraft? I love the old StarCraft. I was really looking forward to the new one until they gutted it by removing LAN play. I would rather play the old version is all its 640x480 glory then play a LAN game over the WAN. Sure I have the bandwidth, but it's the principle. Won't someone think of the Zerglings?
I judt got a nre Kinesis keybiartf so please excusr ant egregiou typos.
Perhaps a game not so dominated by rushing tactics would be a better choice of base game? It definitely seems an interesting idea, but there must be games better suited to an AI contest like this...
Human Advantages:
Advanced Prediction
Flexible Stategies
Arguably Faster Learning
AI Advantages:
Able to command all units at once
Usually More efficient w/ resources
Instant Macro management
Another advantage to the AI could include knowing the map layout and what the player has at all times, which is something the original starcraft had so the AI would know whether to rush you or not.
This is where SkyNet will emerge, :p
Michael
http://s1.sfgame.us/index.php?rec=58163
1.An AI may not injure a terran being or, through inaction, allow a terran being to come to harm. Only Zerg, Protoss and Xel'Naga..
What if, through the developers' virtual arms race, the AIs discover that rushing isn't actually the best way to win? Given enough room to experiment, could new, anti-rush gambits emerge that human players wouldn't have thought of?
Your mind is clear / The things that you fear / Will fade with how much you / Believe what you hear
I hope by AI they mean "smart" not "fast". Anyone who's played around with Warcraft, Starcraft, and the like knows you just can't out build the computer in the beginning. Compy builds in terms of micro or nano seconds; humans, at best, click in terms of half-seconds.
I sometimes don't feel like playing other people, yet the computer is typically a sad joke for strategy beyond the rush. To sum it up, if the new AI is the Zerg Rush master, but too stupid to fend me off should I survive, its fail.
I wonder if the bots will have any restrictions on apm and such. I could imagine the terror of trying to defend against 10 individually controlled mutalisks harrasing all my expansions at the same time, from all directions. And without any long-range trapping mechanism, or fast air-to-air unit, there is no way to catch them. And if I need cannons or turrets or units at all angles of attack, I will certainly lose the macro battle.
Take a look at Sins of a Solar Empire, although it's in space...
The aiide conference web site has been Slashdotted... even though Slashdot didn't link to it. :-)
while (perform_zerg_rush()){ chat('kekekeke ^_^'); }
The source code of the first contestant has been published.
while 1:
t = Tank()
t.attack()
I am sure the SC AI is hard enough to beat already. Do we really need more research in this direction? How about LAN support so I can hunt the forbidden prey: Human?
Winkey shortcut mapping for 64bit windows. WinKeyPlus
Just be aware that a "medium" map game of Sins will take something like 20 hours to play, assuming you steamroll your opposition.
I read the internet for the articles.
...we know when Starcraft II will be released. After October 1st 2010.
I tried to think of a good sig, and this wasn't it.
I just wish everyone would stop calling what we actually have running in today's world with today's commodity computers "artificial intelligence."
With the exception of Fritz or Rybka, there hasn't been a single instance of AI that I personally have been unable to overcome in a matter of seconds, if not minutes. This has lead me to many a dissatisfaction in modern gaming as an adult. As a child, I believe my senses were overwhelmed by the ambiance and graphics, and I didn't think about AI, which I guess made AI work.
I vote for renaming it "Artificial Stupidity." Simply for the reason that it can't learn, and I can. Once I learn it, it is no longer a challenge. It never learns to capitalize on my mistakes, no matter what kind of statistical analysis trickery is happening behind the scenes.
Definitely, brutally... Artificial Stupidity.
"Be prepared, son. That's my motto. Be prepared." --Joe Hallenbeck
What does game AI have in common with traditional AI? Most traditional AI research is focused on reasoning and pattern matching. Adversarial game AI, particularly of the complexity needed for StarCraft, is usually just a greedy algorithm with preprogrammed responses to predetermined scenarios. Don't get me wrong, game AI tournaments are flashy and stuff, which I guess is why MechMania at the UIUC has been going for 15 years, but they have very little in common with traditional AI.
Disclaimer: I only took a couple of AI courses in college, but I am a game developer, as well as a repeat competitor at and attendee of MechMania at the UIUC (which seems like a reasonably close analog to this competition).
AI can 'cheat'.
The fog of war that applies to you, doesn't apply to them. Also, their micromanagement of resources is impossible to coordinate on a human level.
You don't even have to make AI good, just make it cheat. That's how it is good.
What do I win now?
The price is always right if someone else is paying.
Well, I guess it was announced almost 2 months ago. The teams have been submitted and the contest is currently running as far as I can tell FTFA. Hmmm... timely news. I don't think so, this would have been cool back in Sept. so someone that might be interested could simply create a bot and enter it. Now it is way past time
I have mod points and I am not afraid to use them
The cheapest way for the computer to beat the human is to simply open up multiple fronts that cannot be simultaneously micro-managed. Even maintaining one significant skirmish at the front will prevent the human from micromanaging the base back home.
I've been a fan of RTS from the beginning but haven't seen anything really exciting since Total Anihillation. Maybe I've missed some great ideas but pretty much every one I've tried since has been met with the initial glee of pretty graphics and then the crushing disappointment of seeing AI mistakes that were getting old when Command and Conquer first came out.
Honestly, I found Starcraft 1 to be more compelling for the storyline than the gameplay. SCII seems to be the exact same game with prettier graphics.
Kwisatz Haderach
Sell the spice to CHOAM
This Mahdi took Shaddam's Throne
TFA doesn't seem to mention whether or not the submitted AI can cheat and see the entire map. The game's normal AI does this, but humans obviously aren't supposed to. Whether or not they allow this changes the game substantially.
My AI would design its base to be a rough representation of a naughty picture on the minimap. Human players would always lose as they just let the AI build away to see the picture get a higher resolution.
Sins of a Solar Empire is a very good game. Some people complain that it doesn't have a campaign to play through; it's strictly custom games, but that didn't bother me. I sunk days/weeks into it back when I got it and I still go back to play periodically. It doesn't have age advancement exactly, but research is a very important part of the game (to allow colonization of certain types of planets, give new abilities, improve your weapons, etc).
If/when you get sick of the regular game, there are also lots of mods to change things up, including some based on movies/TV series (Star Wars, Battlestar Galactica, etc). Definitely worth it, especially as it gets older and the price drops.
The confidence of ignorance will always overcome the indecision of knowledge.
... I would rather play the old version is all its 640x480 glory then play a LAN game over the WAN. Sure I have the bandwidth, but it's the principle...
So your objection is that you would rather play over a network than over a network.
It looks like the bots must be built using BWAPI, an open source API layer for Starcraft that uses ChaosLauncher, an unofficial Starcraft hack. Seeing as how Blizzard is a super litigious bunch of assholes (read up on bnetd [1], WOW Glider [2], and Scapegaming [3] for a sample) won't they just file a baseless lawsuit against the people responsible for this competition? Given their past behavior, I can't imagine that they would be able to restrain themselves.
<rant>
Honestly, Blizzard makes great games, but they have always fought against free speech, intellectual, and technical freedoms. I hope that Starcraft II falls flat on its face for forcing Internet registration to play the single player game [4] and disabling LAN play [5] in the hopes of increasing profit. The sooner that company fails, the better.
</rant>
[1] http://en.wikipedia.org/wiki/Bnetd#Blizzard_takedown_demand_and_lawsuit
[2] http://en.wikipedia.org/wiki/WOWGlider
[3] http://www.rfcexpress.com/lawsuit.asp?id=51340
[4] This has not yet been announced, but it's naive to think that it won't be the case
[5] http://kotaku.com/5304113/no-lan-play-for-starcraft-ii
I'll buy StarCraft 2 if they have frequent (1 per year or more) competitions like this. Hopefully Blizzard will release its own easy to use API for AI competitions with the release of the game.
The question of whether a computer can think is no more interesting than the question of whether a submarine can swim.
Thats why StarCraft 2 was delayed, they don't even feel like writing the code anymore and want people to do it for free.
"They confiscated everything, even the stuff we didn't steal!"
I imagine that a computer's ability to control units with instant reflexes and frame precision will make AI Starcraft a completely different game from anything we've seen. Watch some Tool Assisted Speedruns and see how the gameplay of a person playing frame by frame transcends that a skilled human playing normally. Games are designed, tested, or balanced with the expectation that a player cannot press a button thirty times a second, anticipate the frame in which a projectile which hit, or issue commands at ten map locations at once. Without these limitations, the game can be broken and become something the designers never intended.
I expect the contestants to abuse lots of bugs and glitches exploitable only with frame-perfect control. For example, there's a known bug when about 1% of the time, a dragoon shot will miss a moving SCV (with no high ground or cover). If something like this can be consistently reproduced, the game will warp. Another rare bug has units become stuck while moving past each other, causing them to dart at ridiculous speeds in a perpendicular direction. This is likely reproducible, and could become the main mode of unit movement in this contest. Even if this doesn't work, there's probably a way to move faster by issuing rapid commands in a way that takes advantage of animations, since Starcraft ground unit movement speed is not hardcoded but animation-based.
Even without bugs, an AI could dance around ranged units to be basically invulnerable to melee, or any slower unit with lower range. This will give Terran a powerful rush strategy (how ironic).
Imagine a game of Terran versus Protoss. The Terran builds a fast barracks, and sends two three marines at the Protoss base. By then, the protoss has two zealots, but they won't matter, since they'll never get a hit. The Terran player dances the marines to shoot the zealots while taking no damage by always moving whichever two marine are being chasing, while the third is free to fire. Even against an equal-sized zealot force, marines are slightly faster that zealots, so they can shoot and move with impunity. The marines can slowly make their way into the Protoss base and behind the mineral line where they'll slowly wear down the mining probes' health, even as the Protoss makes focussing on any one probe impossible. The Protoss might try to get a surround with probes, though I think the marines will escape. Even if the Protoss fends off this rush, the Terran can have vultures before opponents have dragoons and wreak havoc with them.
Since zerglings are faster than marines, I'm not sure if this strategy would work in TvZ, though we might still see some epic bunker rushes.
Note that I don't think this distorted gameplay is bad, just different from human play. I rather like the ridiculous perfection, timing, and bug-abuse of tool-assisted speedruns, and look forward to seeing what the contestants come up with. I would love for a contestant to find a strategy that completely breaks Starcraft as we know it and wins unopposed. However, I think those who expect the final matches to look like really polished high-level human Starcraft play might be disappointed.
Not quite as sharp as your namesake, huh?
YA RLY!
I for one, welcome our Overlord Overlords.
Wait a minute...
Let's teach our AI systems how to do battle... against humans. Skynet anyone?
The robots will be no match for John "Bisu" Connor!
Does this mean they are opensourcing Starcraft or will all those AIs use some kind of lousy plugin api? Anybody know?
Step 1: Playoff/deathmatch style showdown to find the the best of the AIs/human
Step 2: The winning AI plays the 'best' human player
Step 3: Whip (s)he/it to a top class "Ender's Game" facility and make (s)he/it fight for us (afghanistan, Yemen, Secret Darkside Moonbase against the fé'zer - this week only don't ask about next one....)
Step 4: Profit.
Here, I corrected that for you...8)
It takes 40+ muscles to frown, but only four to extend your arm and bitchslap the motherfucker
How is this a troll? I don't think raising the glaringly obvious issue of Blizzard taking legal action against people who hack their games is trolling. We're talking about behavior they have a prior history of engaging in. Anyone who thinks Blizzard is just going to be pleased as punch about this contest is deluding themselves. That's not how Blizzard rolls.