Follow Slashdot stories on Twitter


Forgot your password?
Games Entertainment

Rules-Unknown Artificial Intelligence Competition 176

OOglyDOOde writes: "This link points to a competition being hosted by a company that makes research on artificial intelligence. The task? Build a program that can play a number of games whose rules are totally unknown -- and earn the best score while competing against various opponents. Your program is told the possible choices available, when it should make a move, what did the opponent do; and what was your score for the last turn. There are no entry fees yet there is a cash prize. Submissions can be done in various languages, or in Linux or Windows binaries." This is certainly one of the odder ones I've ever seen, but has interesting prizes (trip to Israel) and rules (fairly broad entry categories).
This discussion has been archived. No new comments can be posted.

Rules-Unknown Artificial Intelligence Competition

Comments Filter:
  • by omnirealm ( 244599 ) on Sunday August 05, 2001 @11:20AM (#2109608) Homepage
    I took a statistics course at BYU. My professor, Dr. Tolley, with the help of his "31337" kids, built an AI system that played Quake. Each possible move was designated as a random variable, and each random variable was weighted according to its success in keeping the player alive and killing the other player. The code would randomly try different actions with the game interface (walk forward, fire weapon, duck, etc.), and then register what worked and what didn't. At first, the computer-controlled player would just stand there. After getting blown apart a few times, it would start jumping to the left, and then ducking, etc. Eventually, it "learned" that it had its greatest chance for survival if it immediately ducked and went behind a box. It then learned to wait until someone walked around a corner, and then it would fire its weapon in the direction of the corner. Finally, it learned that coordinates of the game contained the "respawning" positions, and upon fragging the opponent, it would run to the next respawning point and wait until the player showed up there, blowing him away upon entry into the game.

    This code could be similarly adapted to any game, inasmuch as the code can register a table with all the possible moves provided by the interface. It doesn't even have to know what those moves do; it only needs to know if, by doing certain moves according the "state" (or the attributes) of the game, it gains points (or stays alive or whatever) or loses points. The moves are then given a distribution weighting factor. Then, the algorithm just needs to approximate the game state with the registered table entries, determine which moves have the highest "survival rate" based on the current game attributes, and then perform those moves.

    Depending on the game, it may take a long time before the random variable distribution table gets populated to the point where the algorithm can make "intelligent" decisions, but it works nonetheless.
  • by Anonymous Coward on Sunday August 05, 2001 @01:21PM (#2115475)
    But they don't blow up sinagogues. If they did, and did it consistently, every single day, and if there would no way to single out those who did, then the FBI would go around arresting and if fired at, killing fundamentalist Christians. Which is exactly what IDF does. Palestinians excell at putting children in the front line before the gunmen and then blaiming israelis for using rubber bullets. The international community is fast to blaim israel for bombing a Hammas building because two children were killed as well as two heads of the hammas terrorist organisation, yet when US bombards Yugoslavian cities, it's ok, after all they are protecting peacefull kosovars. Well guess what, Israel is protecting the peacefull citizens of Israel. Our roads are not safe since this october, almost every day more than one shooting incident (that is by palestinians), but you will only hear about it when they succeed in killing more than 10 of us at one time (like the bombing in tel-aviv). But the fact that we are successfull in stoping them before they bomb us is somehow our guilt? I'd like to see you suggest some solution instead of bashing. What do you want us to do? In camp-david we gave it all, and they didn't take, nor suggested any alternatives. In Lebanon we retreated to the exact line the UN resolution requested, UN has confirmed it, yet the hizbollah continues it's attacks on Israel, and since then have kidnapped 3 israely soldiers (from israel's soil) and 1 citizen. It seems that even executing UN resolutions to their letters isn't satisfactory to the other side. Don't forget also, these lands you call occupied where occupied as a result of war that we didn't start nor wanted. In 1948 Israel accepted the division plan by UN, but was invaded by it's neighbour's armies. The result of that was that when war ended, Israel had more land than at the beginning, but arab countries have only themselve to blame for that. Palestinian refugees from 1948 were not accepted by arab countries and still don't have citizenship of the countries they reside in. All this in order to use them as a political card against Israel. Now palestinians want to create their own country, but want their diaspora to be settled in Israel! Don't you think that's a little bit contradictory? What would you said if with the creation of Israel, israel would demand that all countries from which Jews were expelled during the hollaucost (sp) would accept jewish refugees and grant them citizenship? Instead israel has accepted it's diaspora and embraced it within itself. The palestinians are persuing a separatist act, yet they complain that we don't let them in to work in Israel? Do you want independence or don't you? The world demands that Israel stops executing terrorists without trial, yet last week 3 palestinians who were suspected in aiding israel were shot by palestinian authority without trial, and another 3 were sentenced in 10-minutes trial to death. Do you complain about that too? Why don't you? To sum this long rant up: Please, if you don't understand the things you talk about, shut up. The issues at hand are more complex than they come through pictures on BBC or CNN. Also keep in mind, that palestinians don't let journalists to make a free coverage, and most pictures you see are shot by palestinian free-lancers, who despite being journalists, are not impartial, and can and do cut out whatever makes palestinians look bad.
  • by bprotas ( 28569 ) on Sunday August 05, 2001 @12:30PM (#2127669)
    Of course, playing as "badly as they can" implies the same knowledge of the game as playing well, if you want to do better than random chance. To play badly is every much as difficult an AI problem as playing well...
  • by Quixote ( 154172 ) on Sunday August 05, 2001 @07:23AM (#2128039) Homepage Journal
    I think the idea is to make an attemp at "meta-learning". In all of the games that you've mentioned, the programmer knows the rules in advance, and the challenge is to see how best to build a system that navigates through those rules. In this contest, the idea is to see how you can capture the "programmers' thinking".
  • random fortune... (Score:5, Insightful)

    by Pelam ( 41604 ) on Sunday August 05, 2001 @04:33AM (#2128459)
    Probably irrelevant, but a fortune came to my mind when reading that submission:
    In the days when Sussman was a novice Minsky once came to him as he sat hacking at the PDP-6.

    "What are you doing?", asked Minsky.
    "I am training a randomly wired neural net to play Tic-Tac-Toe."
    "Why is the net wired randomly?", inquired Minsky.
    "I do not want it to have any preconceptions of how to play".
    At this Minsky shut his eyes, and Sussman asked his teacher "Why do you close your eyes?"
    "So that the room will be empty."
    At that momment, Sussman was enlightened.
  • Re:MOD UP! (Score:4, Insightful)

    by MOMOCROME ( 207697 ) <> on Sunday August 05, 2001 @08:09AM (#2131178)
    I second that. This is a reference to a Koan found in 'Escher, Goedel, Bach, an Eternal Golden Braid' by innimitable Douglas Hoffsteder/

    Otherwise known as the seminal work of AI philosophy.

    This is truly on topic, moreso that the un-enlightened could ever know. ask yourelf: Are my mod points the mod points of the un-enlightened? if no, please mod up the parent's parent as +1, Insightful.

  • by Vryl ( 31994 ) on Sunday August 05, 2001 @05:43AM (#2133946) Journal
    It's not like the entries have to be intelligent. They just have to be logical and well designed, and good at pattern recognition.

    All depends on the much debated definition of what is 'Intelligence'.

    Certainly, pattern recognition is a sign (symptom?) of intelligence.

    So, what are you actually saying? What do you mean when you say 'intelligence' ?

  • by tardibear ( 135254 ) on Sunday August 05, 2001 @05:55AM (#2134326)

    Heck, I'd like to see a competition where HUMANS play a game where they don't know the rules. That could be just as intereting.

    You're playing it right now, bud ..

  • by theneo ( 511389 ) on Sunday August 05, 2001 @04:32AM (#2135831)
    Not a programming marvel IMO. Could easily be very hard to pull off, but the level of programmin g isnt revolutionary. What really sticks out to me is this is a totally different way to look at AI.

    If its pulled off, should revolutionize as to what as seen possible for AI programming. ChatBots have been around for ages. That sort of AI exchange just isnt impressive. If this is done, at least in my eyes, it will prove points and open doors for future projects. Should be interesting.
  • Two thoughts (Score:4, Insightful)

    by Cave Dweller ( 470644 ) on Sunday August 05, 2001 @04:42AM (#2137639)
    1. If I win, do I get a trip to the US? (see email address)
    2. I wonder whether the winner could visit me.

  • by slasho81 ( 455509 ) on Sunday August 05, 2001 @07:47AM (#2138369)
    After reviewing the challenge rules, I see this challenge as a simple exercise in neural networks coding.

    The challenge is so obscure that any entry submitted will have to deploy a very generic NN and a trainer. this basicly means that after enough training any entry would do sufficiently good at any simple game (such as scissors, rock, paper) but playing anything more complex than
    that is shooting in the dark. The interface and the rules of the challenge themselves are too obscure.

    If there is someone with a code that could win such uncertainty effectively and efficiently, he'd be stupid to submit it for $2000.

    Then again I must give a person that can do something extraordinary as that some credit for not doing something that stupid.
  • by A nonymous Coward ( 7548 ) on Sunday August 05, 2001 @10:54AM (#2150086)
    They are playing the STOCK MARKET. They buy stocks according to the various submissions, gradually weed out the bad performers, and end up making a pile, with which they can pay the prize and still have a tidy profit.

    Wish I'd thought of it!

Truth is free, but information costs.