Slashdot Mirror


Rules-Unknown Artificial Intelligence Competition

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).

176 comments

  1. My Statistics Professor Has Already Done This by omnirealm · · Score: 5, Insightful

    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.

    --
    An unjust law is no law at all. - St. Augustine
    1. Re:My Statistics Professor Has Already Done This by Anonymous Coward · · Score: 0

      You're correct, that's what has to be done, and certainly it's what most of the programs taking part in the contest will be doing. So, the winner will be, probably, the program which learns faster than the other programs. So, when slower programs figure out some strategy (whatever it may be), it can adapt itself, change its moves and come with a new strategy to win. Of course, this is not an easy task. A program that is able to learn can be done easily, given time and memory enough. A program that is able to learn faster than all the others - that's hard.

  2. Re:some design specs for potential participants by Anonymous Coward · · Score: 0

    I would say... ask jeeves.

  3. Re:random fortune... by Anonymous Coward · · Score: 1

    And this is one of the reasons Minsky set back research on neural nets for 20 years...

  4. How many actual AIs reading slashdot? by Anonymous Coward · · Score: 0, Funny

    How many actual AIs reading slashdot? And feeling superior?

    1. Re:How many actual AIs reading slashdot? by Anonymous Coward · · Score: 0

      please go stand by the stairs.

  5. Re:Even if I could write this program i wouldn't.. by greenrd · · Score: 1
    Fundamentalist Christians don't like Jews much either (or indeed anyone who doesn't agree with their religion), but Jews in the US don't use that as an excuse to go around killing fundamentalist Christians.

    Just because someone hates you idelogically, that's no reason to treat them like shit.

  6. Re:Offtopic: Trip to Israel (not really offtopic) by WinterKnight · · Score: 1

    Bla bla. Politics. Bla.

    Anyway, as an Israeli I can assure you that the
    "situation" has some really low chances of
    hurting any tourist. Fact: i'm taking the bus
    on a daily basis and yet i'm still alive. And
    I -live- here. So, really. The media just like
    to over bloat things.

    On the down side: Its freaking HOT. Dont get
    here unless you're heat tolerant. I'd take
    a trip to swiss instead at any time.

  7. Even if I could write this program i wouldn't... by Anonymous Coward · · Score: 1, Flamebait

    ...for all I know the crazy Israeli army/government will use the code to write programs to further their inhumane agenda. Think about it for a second. Where should we move our tanks, guns, planes and other US Tax payer supported weapons to target the Palestinians.

    Going through UN archives or resolutions, even the United Nations condemns Israel for taking the land by force the way they did. And what's more ridiculous, is that now they say "You palestinian's shouldn't fight us, we're ready to make peace on our Israeli terms". Ridiculous! That's like me walking into Taco's home and kicking his butt out and then when he fights back, I cry to all of Slashdot saying he is psychotic. And when that's over, I tell him that he can live in the mulch pile in the back which I sleep in his room.

    I know this message will get moderated down to -1, and is it doesn't people will sit here an attack me. Both sides have done wrong, I just think the people who started it (Israeli's) should take their stuff and walk away. Didn't the holocaust help them realize anything? What's ironic is that the holocaust of 21st century is being conducted by the Israeli's and our US taxpayer money supports them with weapons!

    Ultimate lameness...!

  8. Re:The game is Slashdot, the score is Karma. by Steeltoe · · Score: 1

    Look at chess as an example. This is like having a chess computer that has to learn the rules. Compared to playing chess (which is computable), learning the moves is relatively easy.

    Wrong. Let's suppose this program gets a list of available moves every turn. How is it to even know the board is a 8x8 board and the relations between each square? How is it supposed to know the significance and difference between a king, queen, rook, knight, bishop and pawn? In order to play intelligently, it will have to learn that sacrifices can be a good thing, but is generally bad if you aren't guaranteed an advantage later in the game. That making one little mistake in your tactic/strategy can quickly lose you the game against a strong player.

    Without assumptions you're really screwed. So it very much depends on the interface, how much and what information the program gets every turn. There'll always be some assumptions left too, as what you choose will always have some limitations.

    - Steeltoe

  9. Re:That recall me a couple of clever hack... by Anonymous Coward · · Score: 0

    s/the one that did not/the ones that did not

    Cheers,

    --fred

  10. Re:This should prove entertaining. by krogoth · · Score: 1

    "I'll move my piece to square A17..."

    :*BEEP* Sorry, but you lost!"

    --

    They that quote Benjamin Franklin on liberty and safety deserve neither.
  11. Re:Even if I could write this program i wouldn't.. by Com2Kid · · Score: 1

    Oh great, how lovely, an almost compleatly useless set of additional integer CPU instructions, and another product made to be AOLs biatch.

  12. Re:hmm by CaseStudy · · Score: 1

    GA won't win. If it's done beforehand, it will optimize itself for specific games. If it's done afterward, it will fall behind early and be unable to catch up.

  13. some design specs for potential participants by beanerspace · · Score: 4, Informative
    1. Re:some design specs for potential participants by sander123 · · Score: 1
      A very interessting collection of lists!

      I just finished reading the second and it is a thought provocing article.

  14. Some black humor by Anonymous Coward · · Score: 1, Funny

    If you win, you can win a trip under the bomb in israel. And if you die in this second game, they keep the source.

  15. Re:Offtopic: Trip to Israel (not really offtopic) by Betcour · · Score: 1

    Well I'm sure some places in Israelien territories are very fine and rather safe - but for one many interesting places are in Palestinian territories (and no tourist wants to be taken in a crossfire), and for two I would really hate to provide, with the money I would spend, support to a country who is organizing etchnic cleansing and conquest thru colonization. The same as I would not visit North Korea or Birmania.

  16. How will they do this? by Hank+the+Lion · · Score: 1

    From the detailed information on their web-site:
    A round-robin tournament will be held to select the winner of the Learning Machine Challenge. All combinations of players will take part in all games, of which there will be between six and twelve.

    As I see it, they plan to let every contestant play against every other, on 6-12 games, several thousand moves each.

    Where will they find the time to do this if they get more than just a few dozens of entries?

    1. Re:How will they do this? by Anonymous Coward · · Score: 0

      7?

    2. Re:How will they do this? by dlkf · · Score: 1

      There will probably be fewer than 100 contestants(a typical machine learning conferrence will draw only a few hundred participants including those presenting and I doubt this would atract more attention), and more likely around 20. At 100 contestants, using your other estimates, the time drops to about 114 computer days (at 20 contestants, it drops to about 4 computer days). With 10 computers, thats less than a week. Not unreasonable for something like this.

    3. Re:How will they do this? by JM_the_Great · · Score: 1

      It'll be done by a computer program - hence the standard console interface.

      --

      --Justin Mitchell
      "2nd Place is a fancy word for losing" --Bender (Futurama)
    4. Re:How will they do this? by Anonymous Coward · · Score: 0

      How many people do you think:

      a) are interested in AI and game theory?
      b) have some decent programming skills?
      c) have enough time before Oct. 30th to do this?

    5. Re:How will they do this? by Hank+the+Lion · · Score: 1

      It'll be done by a computer program - hence the standard console interface.

      I realized that.

      But, let's say we get 1000 contestants.
      Also, they specify 6-12 different games for each, let's say 10 on average.

      This will mean 10*1000 * 999 / 2 matches (when A has played against B, B has played against A) of many thousand moves (let's say just 2000) each.

      A program is required to process at least 10 moves per second. Worst case, this will lead to roughly 1000 million seconds, or 30 years, of computer time.

      When you get more than 1000 contestants, the time required for the match will rise quadratically.

      I realize that the average program will be faster than 10 moves/second, and that you can use several computers to speed things up (from the height of the prize, I gather that their budget is not unlimited, so I think more than 10 computers will be out of the question), but still, if you get a significant number of contestants, letting every contestant play agains very other may be prohibitive.

    6. Re:How will they do this? by wheany · · Score: 1
      7?
      It was a rhetorical question...
  17. Re:Even if I could write this program i wouldn't.. by WinterKnight · · Score: 1

    Last I checked, democracy isnt perfect.

    There are many other factors for why people choose who they choose. Keep looking at the shallow side, and you'll never get anywhere.

    Not that you tried to, anyway.

    PS:
    Incase you didnt notice, we didnt fire a single shot against those "poor, innocent" people for months. And yet, they keep on shooting. When we do shoot, its rare, and in response to something they did. Very inhuman. I guess that we should stop responding at all and just sit back and get killed. Fun, isnt it.

    And dont start with the "who did what first". When you get killed, it doesnt matter. When we tried to talk, they didnt stop shooting for a moment. During the peace process, they kept saying that "peace is just a cover! We will kill them all anyway!" in public radio and TV. But, its ok. They're palestinians. They are permitted to do it.

  18. Interesting by Johnny+Starrock · · Score: 1

    I don't claim to be an AI guru, but don't programs/computers live and die by rules? (Bad example: HAL in 2001. Rule: Complete Mission. Obstruction: Humans.) The AI could only adapt so much...

    --

    end communication
    1. Re:Interesting by pmcneill · · Score: 1
      Not necessarily. There are rule based systems ("when I see X, I do Y"), and these can fall apart in unexpected situations. However, this is far from the only way to implement a system.

      One method is with classifier systems, where you evolve the rules that determine the output based either directly on the input, or a chain leading from some input. It starts with a pool of random bit strings which are evolved based on their success. The rule used is determined by a bidding scheme.

      Another method, which is about as general as you can get, is genetic programming (GP). GP involves creating a set of functions and terminals and randomly generating a set of parse trees using them. Each of these programs is evaluated, and based on that the standard genetic algorithm operations are performed for form a new generation. Essentially, genetic programming is automatic programming, if given the right function and terminal set. Unfortunately, it would probably be too slow a process for this competition.

      Both of the above methods have been proved over and over again. Classifier systems, for instance, have been used to run a simulated oil pipeline (with leaks, blockages, etc). Starting from a random population, it achieved human competitive results. Genetic programming has produced results that are not only human-competitive, but also infringe on pre-existing patents.

  19. Re:I might know how to win or get an unfair advant by CyberDruid · · Score: 1

    The point is not to decline playing against the other real programs, just for the fake progs to play to the best of their ability against everyone but you.

    Sending ID will most likely mean making a commitment, but for long and somewhat stochastic games (which I have a feeling that we will be dealing with here), this early commitment will (hopefully) not be enough of a handicap for the fake progs. They will still get points (provided that your algorithm is any good in the first place).

    The real program will only reply an ID string when it is statistically sure that its opponent is a "buddy", thus the real program gets the unfair advantage of getting a few/many easy wins that the programs made by the other contestants will not get, but sacrifices nothing. You will still need a good program to win. This is just how one could (if one were so inclined ;) give it that extra boost.

    //David Fendrich, Swedish AI-dude
    --

    Opinions stated are mine and do not reflect those of the Illuminati

  20. Re:This should prove entertaining. by lavaforge · · Score: 1

    If you're an American, it's called "cricket."

  21. Re:Easier than I feared by KFury · · Score: 2

    But with a game as complex as chess, for example, there is no mathematical way that the program could gather enough information over the course of a single game to make that kind of assessment.

    Lose a queen and you never know what it is to not have lost a queen, for example.

  22. Israel, Prize? by Anonymous Coward · · Score: 0

    A trip to Israel is a prize? Sheesh, they don't even allow tourists there right now.

  23. Re:The game is Slashdot, the score is Karma. by rking · · Score: 1

    something is intelligent when another intelligent being (i.e. a human) cannot tell if it's a machine he's communicating with, or another human

    So for example, you aren't intelligent because I'm confident that you're a human not a machine?

  24. Try Mornington Crescent. by TDScott · · Score: 2

    A fantastic game for experienced players, but newcomers can sometimes be confused at the start.

    It's played on the stations of the London Underground network. For example, I could start with Albright's Opening, *Regent Street*.

    Anyone want to take me on?

    1. Re:Try Mornington Crescent. by TDScott · · Score: 2

      Well, Scott-Berkely's more suited for online play: subtlety really comes out over the 'net.

      I'll go for Ealing Broadway (hence declaring a state of emergency).

    2. Re:Try Mornington Crescent. by Konovalev · · Score: 1

      Ah, I see you are playing Scott-Berkely Rules. Most MC players over here (IMHO unjustly) regard that as too simplistic, but I reckon there are possibilities for some fairly subtle tactics. I take it redoubling is allowed (Scaife's Variation excepted, of course)?
      In that case...St John's Wood.

  25. Re:Easier than I feared by greenrd · · Score: 1
    Your program can attempt to make theories about what moves give you what scores in what contexts, test those theories, and if they seem to work, continue with them; if not, try another theory. The program has to be a good scientist - working out the rules of the world in which it finds itself.

  26. The simple solution by dozing · · Score: 1

    I'm going for the easy answer: If my score is an even number the answer is yes. Otherwise my answer is no.

    --
    Dozings.com -- Its kinda funny... If you're as crazy as me.
  27. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    I think you need to read some UN resolution which directly condemn israel for taking palestinian land...

    One of a hundred resolutions here

    There are hundreds more. The original poster is most likely claiming what the rest of the world and the United Nations all know and believe. How was Israel created in your opinion? The land was empty and they walked in? Lets hear it so we can all point you to the historical documents you need to read so that you understand that Israel not only invaded the land but they murdered for it.

  28. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    I agree with the orginial poster and his/her opinions. I would have probably written a harsher and therefore truer to reality statement against the Israeli government and IDF. My entire family was ruthlessly murdered by them.

  29. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 1, Insightful

    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.

  30. Re:humans? by Anonymous Coward · · Score: 1, Funny

    True! Or, even better - bacteria!

    I remember that some few years ago some guys used bacteria to predict the stock market. apparently they are more susceptible to the miniscule trends humans tend to overlook. They really know how to adapt to beneficial environments, and, accordingly beat many an analyst.

    However, with a doubling rate of over 20 minutes, they won't have a chance in Quake... Not even against Joe Schmoe with a 28.8k modem!

    However in a slow, perhaps turnbased system they could be killers.

    Sensmoral - that gaming competition has little to do with TrueIntelligence(tm). If one gets beaten by a gang of Streptococcus sp. it says very little indeed, but more perhaps on stock market analysts. ;)

  31. Even more interesting... by Cpt_Corelli · · Score: 1

    ...was the other information on the AI site about how they are working on the Child Machine HAL. Can you teach them how to do Java? In that case I could "work" from home more often...

  32. User-submitted binaries? by whatnotever · · Score: 1

    Will these programs be run in a sandbox? They're accepting user-submitted binaries, so I would hope so. To be completely fair, the submitted programs should be able to do nothing but read and write from stdin/stdout, otherwise they may do any number of things. Even if they are restricted like that, the judging program had better not having any buffer overflow vulnerabilities. ;-)

    Somehow, I don't get the feeling that these people have planned this very thoroughly. There are other little things that don't quite seem right, too...

    1. Re:User-submitted binaries? by robbieduncan · · Score: 1

      The rules state that the programs are not allowed to open network connections (so you can't get more processing power I assume). Because of this they are able to state on their site that all programs will be run on a machine isolated from the network.

  33. Re:Easier than I feared by SuperKendall · · Score: 2

    Try reading the post.

    Imagine in chess where you scrifice a pawn to gain position - in that case your "score" that turn would be negative yet you would be able to gain a higher overall score later as a result. With only knowledge of the "score" for that turn, how can you decide when a sacrificial move might leave you better off?

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
  34. Re:I might know how to win or get an unfair advant by e-gold · · Score: 1

    Interesting idea, but I don't think it would be unfair, since according to the rules you can enter as many times as you want (don't even bother with friends, unless you want to). Also, I got the feeling from reading the site that programs are pitted against eachother one on one (I don't know for sure, perhaps this is another game rule that could change from game to game?). If this is the case, your "real" program might be subjected to other competitors' real programs and be unable to decline playing with them. Also, the judge program might not allow any identity information through until after your program makes a commitment. Still, assuming enough of the "fake" entries, it might still be possible to gain an advantage by dint of sheer numbers if you entered, say, a million times.

    I suspect, if they ever try this contest again, they'll want to ponder that rule a bit more...
    JMR

    --
    Try e-gold - (contact me). I'm NOT e-
  35. Re:But the real question is by Anonymous Coward · · Score: 0

    But the real quetion is, out of 100 some odd links, these five or six were interesting ? I think so !

  36. Re:Are you that arrogant? by Anonymous Coward · · Score: 0

    Yeh, the Israeli's really struggled in that 6 day waar thing. Only the worst luck could ever have defeated allah's chosen.

    And this "world" you claim knows Israel's at fault, if it's the same world I see and know every day, then fuck it. I don't care for its opinion on anything else either.

    Arafat's speechs are in Cairo's newspapers every so often, and presumably in other arab cities also. Maybe the translation was lacking, but even so he made it pretty clear what he believes the future has in store for the israeli's. Only an incredibly biased world media would ignore than, then publish stories about how the Israeli goverment poisons arab well water, or uses palestinian babies for vivisection experiments. The palestinian propaganda machine doesn't even bother to make up plausible lies.

    And as far as Islam being at fault, I've never said that. Just some twisted version of it practiced by people that are hardly even human by my standards anymore.

    I'll even go you one farther... I bet there are palestinians who just want it to stop, and live peacefully. Of course, when they aren't being murdered by their own coutnrymen as evil jew sympathizers, they are probably ignored by the "palestinian goverment". Guys, does anyone know if this is the first official goverment to have a known terrorist as its chief executive officer?

  37. Mid East Solution (OT) by Tablizer · · Score: 0

    I don't understand why both sides won't simply agree that for every isreali killed by terrorist attacks, X new settlements go up (and lessor points for limbs, etc).

    This seems like a self-correcting solution rather than the decade-long "no, you stop first".

    At least TRY it. Nothing else is working.

  38. Re:This should prove entertaining. by SimCash · · Score: 1
    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.

    It is called life! (Now, where did I leave the rule sheet?)

  39. Re:rand() by dlkf · · Score: 1

    That is because some amount of randomness is necessary to get intelligent search algorithms out of local minima. This is why genetic algorithms and simulated annealing, for example, have seen success.

  40. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    No one has the right to murder innocents to promote a political agenda. The israelis are hardly guilty of invading the land, as the original poster suggested.

    You may be right about the history, no idea. Maybe if I don't learn it, I'll be doomed to relive that peace.

    The particular flavor of Islam that arafat and his cronies preach, tells them to kill their enemy, and that only by doing that can they reach heaven.(Actually, it says to genocide the enemy, leave not one alive. Look it up, if you don't believe me) For whatever reason, they picked the israelis to be that enemy. Proximity=convenience? Bad choice IMO though... Generally, when picking a fight, you don't insult someone that can obviously kick your ass.

  41. Re:This should prove entertaining. by Anonymous Coward · · Score: 0

    Right on spot.

    Cheers,

    --fred

  42. Easier than I feared by KFury · · Score: 3, Interesting

    The first question that popped into my head was "How do we know what the opponents move means?"

    I was under the misconception that at each turn in the game, the judge will inform the player of all possible moves (as in chess, checkers, or the like) but looking at the specification, it seems that the moves are detailed at the outset of the game, and then are available to each player at each stage in the game.

    now the odd thing to me is the measure of 'state' in the game. Is the score that's returned after each move the current cumulative score, the score for that move alone, or what? Also, what is the goal of the game? It would be short-sighted to assume it's to amass the highest score. In effect, the score is just another input variable, along with the opponents move, which may or may not be useful for judging what is a good move or a bad move.

    For example, if you were trying to make an algorithm to solve the A8 puzzle (the 'sliding tile puzzle' with 15 tiles and 16 spots), and the computer judged your score by totalling up manhattan distances to the goal state, that may or may not be a fair scale of how many moves away you are from winning in an ideal case.

    The system is still underspecified. Without knowing what 'score' means, and whether it is an estimate or a deterministic function, then the project is pretty much a game of luck, and coding is not an effort of skill.

    1. Re:Easier than I feared by Anonymous Coward · · Score: 1, Informative

      Try reading the specs.

      "Score" is the number of points scored that turn. It is a floating point number between -1 and 1.

      The goal *is* to get the highest score.

    2. Re:Easier than I feared by dlkf · · Score: 1
      "How do we know what the opponents move means?"

      It really doesnt matter so long as the names for the all legal moves are given to you at the beginning. Just think of them as symbols, buttons to push at any given time. You shouldnt be trying to interpret meaning in the moves that are made, only value. It will be impossible to create some kind of high level of understanding about what a move means because the program does not know what the game is. It is possible to learn what values moves have though using some kind of reinforcement learning algorithm. Think of it this way: If your program believes it is in a certain state and its opponent executes a certain move, it should pick the move available to it that maximizes the expected reward

      Is the score that's returned after each move the current cumulative score, the score for that move alone, or what?

      Based on the example used in the specs, I would imagine that the score is for the most recent action if given after each action or cumulative if given only once at the very end. Either way, it will be relevant in the global context of the entire contest since the program with the highest overall score after playing all games against all opponents will be the winner.

      The system is still underspecified. Without knowing what 'score' means, and whether it is an estimate or a deterministic function

      I think its better to think of score like a reward or a penalty. The idea is to accumulate the most reward or the least penalty. As they say in the specs, the program with the highest score is the winner. If chess were to be one of the games(which I doubt it will due to its complexity), the score would probably be 1 for a win, 0.5 for a draw and 0 for a loss, just like at any chess tournament. There would be no intermediate score for taking a queen or getting into a strong position because those things are meaningless after the game is finished.

      coding is not an effort of skill.

      This contest is not about coding, its about learning. They dont care how good you are at writing programs, only at how good you are at designing a robust system that learns quickly. And as I hinted at before, this problem is just screaming for contestants to use a reinforcement learning algorithm.

    3. Re:Easier than I feared by dlkf · · Score: 1

      As they mention in the specs, you will play thousands of rounds for each game. I dont think that they mean that each game will consist of thousands of moves (even with random move selection for both players, chess usually terminates after 200-300 moves), I think they mean that you will play each game thousands of times against each opponent. If you only play a game once, very little learning will be done. This is because all learning requires either instruction or exploration. Since there is no instruction in this contest, all learning must come from exploration and one game would not be enough.

    4. Re:Easier than I feared by Anonymous Coward · · Score: 0

      Read the rules. You get to play multiple games of a particular rule set. So, you're right, one game isn't enough, but they know that.

    5. Re:Easier than I feared by Anonymous Coward · · Score: 0

      Actually, they're right. I had to read through the pages a couple of times before I found it, but the overall winner will have the highest total points across the entire tournament. It's simple, but, somehow, unsatisfying. I like the idea (however implausible) of somehow getting your program to recognize when it should be strategic, and when it should be tactical, but this one looks like (more or less) straight tactics.

      And, though I probably haven't thought about it enough, I feel like the restriction that all moves be available on every turn is probably going to eliminate most of the interesting cases. Even tic-tac-toe falls to that requirement.

      Still...

  43. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    Dufus, the same UN you are using to promote your agenda. Guess the UN was evil way back when, but now it's dandy. Funny, you imply that it's just great, the same year the human rights council is packed with nations like china, north korea, etc, and the USA is kciked off it.

  44. Calvinball by bravehamster · · Score: 5, Funny
    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.

    It's called Calvinball, and it's the sport of kings.

    --
    ---- El diablo esta en mis pantalones! Mire, mire!
    1. Re:Calvinball by beowulf_26 · · Score: 1

      Close but not quite. In Calvinball the rules are allowed to change, whereas I think everyone else was speaking in terms of a set of predetermined, yet unknown rules. Bascially, the contest is a test to see how fast/well a computer can learn the rules to the game. Although, I think a good followup to this "experiment" would be to create programs that play Calvinball and constantly report the known rules.

      Here are the rules of Calvinball

      --

      --I hate big sigs.
    2. Re:Calvinball by Anonymous Coward · · Score: 0

      *Strikes you with the Mallet O' Correction*

  45. Re:I might know how to win or get an unfair advant by Anonymous Coward · · Score: 0

    Such groundbreaking AI theories. You should use this for your Phd dissertation... at Suck University.

  46. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    c'mon.. palestinians have every right to fight back.. it was their land in the first place before israelis came over with tanks.. and history has it that jews,muslims and christians actually lived together PEACEFULLY before the crusades in jerusalem.. i don't think there is any such clause in any religion that tells you to kill thy neighbour..

  47. Re:hmm by dlkf · · Score: 1

    Actually, this sounds more appropriate for some sort of multi-agent reinforcement learning algorithm. Anytime someone mentions actions, rewards and learning in the same sentance, the first thing that comes to my mind is reinforcement learning.

  48. Re:I might know how to win or get an unfair advant by glyph42 · · Score: 1

    You can't send arbitrary strings. Read the webpage. You can only look at moves, and make moves when your asked, out of the list of legal moves you're given.

    But you could adapt your idea to this situation. Simply have your "master" program know the internals of all your "slave" programs, so that it can identify a likely slave based on the first few moves. Since it knows the slave's internals, it can beat it 100% if it's actually a slave. If it deviates from the known slave behaviour, well then you need a real algorithm.

    In this scenario, your slaves don't even have to suck. They can use the best algo you can muster, but the master prog will still beat them because it knows what they're thinking. You can even use the same algo in the master when it identifies a real opponent.

    Dead simple to do this if you already have a half-decent algo. Someone want to try it? Easy 2K...

    --
    Music speeds up when you yawn, but does not change pitch.
  49. Offtopic: Trip to Israel by absurd_spork · · Score: 5, Funny
    Sorry for posting off topic, but I'm not sure if a trip to Israel is that desirable as a prize at the moment, given the rather unstable situation there.

    Of course, there may be some connection between the prize and the game ("win a conflict where you've got no clue of the rules", that pretty much sums up the problems of both parties in the Middle East).

    1. Re:Offtopic: Trip to Israel by Anonymous Coward · · Score: 0

      > Sorry for posting off topic, but I'm not sure if a trip to Israel is that desirable as a prize at the moment, given the rather unstable situation there.

      OTOH, the flight prices are dirt cheap...

    2. Re:Offtopic: Trip to Israel by Anonymous Coward · · Score: 0

      The positive side is that any homocidal maniac can go to israel and fulfill their killing fantasies on the palestinians and the israelis will consider them heroes.

    3. Re:Offtopic: Trip to Israel by arikb · · Score: 1
      When you DO go to Israel, it's helpful to remember:

      You are more likely to be killed by a car accident than by a terrorist attack, with the odds being 5:1 against.

      Yes, there are terrorist attacks. Yes, they kill us, and we kill them. Yes, we live our daily lives, and so do they. Don't panic, folks.

    4. Re:Offtopic: Trip to Israel by Anonymous Coward · · Score: 0

      > Yes, there are terrorist attacks. Yes, they kill us, and we kill them. Yes, we live our daily lives, and so do they. Don't panic, folks.

      You're palestinian, that's it ?

    5. Re:Offtopic: Trip to Israel by Bob+Uhl · · Score: 2
      Heh. That and the fact that I'd rather get a trip to Palestine. Spend my tourism dollars where they go to support a government not bent on genocide.

      Still, who wants to go to that part of the world right now? Is there something about landing in the middle of a shooting-blowing-people-up war that that just says `must-do vacation' to some folks?

      I do like your idea about it being to play out the Middle East peace process. Couldn't be any worse than any of the current players.

    6. Re:Offtopic: Trip to Israel by Anonymous Coward · · Score: 0

      Oh please. Do you really think that killing terrorists is a bad thing? After all, they're the only people the Israelis are trying to kill. All the conflicts that didn't have to do with terrorists were started by the palestinians. Oh, and dare I mention the $10,000 reward Saddam is paying to the family of every "martyred" child?

    7. Re:Offtopic: Trip to Israel by Bob+Uhl · · Score: 2
      After all, they're the only people the Israelis are trying to kill.

      Whose land are the Israelis camped on? Whom did they displace? Who are second-class citizens on their own real estate? Here's a hint: while it is a Semitic people, it sure as shootin' ain't the Israelis. They stole that land, and the Palestinians want it back. We have our Indian issues--at least we paid lip service to the Indians and their nations. The Israelis simply took the land they were given. Note also that the Israelis are in violation fo the fourth Geneva convention with their settlements of conquered territory. Note also that Sharon is shortly to be indicted for various crimes against humanity.

      Oh, and dare I mention the $10,000 reward Saddam is paying to the family of every "martyred" child?

      Good for him. What's your point?

      My point was a) I'd rather not visit that part of the world b) I'd rather not give any money to the Israelis c) if I had to give money to any of 'em, it might as well be the Palestinians, who have at least a legitimate beef. Not that I care for them that much either.

  50. Re:How many actual AI researchers reading slashdot by Anonymous Coward · · Score: 0
    THANK you. I'm a cog. sci grad student. I don't know why I read the AI threads; they always piss me off when people comment who don't know anything about anything except having read Neuromancer and taken one undergrad AI course whose most complex AI feat is a minimax checkers player.

    chump: AI hasn't progressed in 50 years. Its a failure.
    me: for starters: Backprop, RBF, HMM, SVM, SOFM, ART, plus tons of symbolic side that I don't know about, people are mapping neuronal layers in the auditory and vision systems. Just because there isn't "strong AI" yet doesn't mean the field has failed. Physics doesn't have a grand unified theory, medicine can't make people live forever, etc. & we don't consider those fields failures. NNets are being moved out of the research labs and into everyday products, much like computing technology not too long ago.

    chump: AI won't go anywhere until they acknowledge that they need (insert random physical property of the brain hypothesized in a pop science book or documentary)
    me1: Penrose should stick to physics
    me2: Tilden's analog/BEAM is a 3rd rate neural network implemented in hardware. Its incapable of doing anything more than oscillations. Analog hardware is irrelevant; modern computers are quite capable of doing floating point arithmetic.

  51. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0


    News Flash!

    The USA goverment, working in concert with other evil nations, invaded poor little Germany in 1945. That they may have had just a little bit of cause, doesn't change the fact that they are evil.

  52. Nag-A-Tron! by Tablizer · · Score: 0

    Re: Give me a woman's brain anyday!

  53. Re:Can it be? by Anonymous Coward · · Score: 0

    i think not

  54. MOD UP! by Bahumat · · Score: 1

    Mod that one up folks...

    --
    "To pass through the jungle; silence, courtesy, ferocity, as the occasion demands." -- Kamau, "Proper Passage"
    1. Re:MOD UP! by MOMOCROME · · Score: 4, Insightful

      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.

      thankyou.

  55. The best hack Re: of clever hack... by Anonymous Coward · · Score: 0
    The rules say nothing about not doing anything under water. So it is a gues that there runs an other program that communicates with in on the console. Manipulating this program, or the opponanan with some clever (fork?) code will make you win.

    Cook the karma

    There is NO FORK

  56. Re:rand() by matek · · Score: 1

    Heh... Some years ago I was involved in some competitions in Denmark and Europe for high-school students ( programming ). Some challanges involed a bit AI knowledge - but each and everyone of the winning entries have used a random algorithm to solve the challanges, and still always had biggest success rate...

    World is weird....

  57. Re:I might know how to win or get an unfair advant by glyph42 · · Score: 1

    Oh, and I just remembered that the judge program can insert moves that are part of the game, but not initiated by your opponent. This means you don't know if the opponent made the move or if the judge did it. So you'll need some fuzzy matching on your identification algorithms... hmm. Maybe not so easy.

    --
    Music speeds up when you yawn, but does not change pitch.
  58. Don't do it by Anonymous Coward · · Score: 0

    It's just a trap to get more AI people into Israel Isn't this kind of setup used to nab probation violators? "You've won a free prize, come to this location to pick it up."

  59. Re:How many actual AI researchers reading slashdot by Mister+G · · Score: 1

    at least one, chief!

  60. The game is Slashdot, the score is Karma. by Basje · · Score: 2

    I wonder what will happen if you let one of these entries loose on slashdot? Would you get intelligent posts, or intelligent trolls? Imagine such a program accumulating more karma than Jon Katz. That would be a boost :)

    But seriously. How can one consider this contest artificial intelligence? It's not like the entries have to be intelligent. They just have to be logical and well designed, and good at pattern recognition.

    Look at chess as an example. This is like having a chess computer that has to learn the rules. Compared to playing chess (which is computable), learning the moves is relatively easy.

    --
    the pun is mightier than the sword
    1. Re:The game is Slashdot, the score is Karma. by Basje · · Score: 1

      I mean the Turing definition: something is intelligent when another intelligent being (i.e. a human) cannot tell if it's a machine he's communicating with, or another human

      --
      the pun is mightier than the sword
    2. Re:The game is Slashdot, the score is Karma. by Vryl · · Score: 2
      Let me see ... Turing NEVER defined intelligence.

      Turing said the following, in his famous paper "Computing Machinery and Intelligence":

      "I shall replace the question [can machines think] by another ..." and goes on to describe the 'Imitation Game', which is so often now mis-labeled 'The Turing Test'.

      He then goes on to say, after describing the imitation game;

      "We now ask the question, 'What will happen when a machine takes the part of A in this game?' Will the interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman?"

      The problem these days is that when the Imitation Game is played, the interrogator has sometimes picked humans as machines. So, basically, the Imitation Game is worthless as an indicator of intelligence.

      If it can't tell humans from humans then it is useless to try and tell humans from machines.

    3. Re:The game is Slashdot, the score is Karma. by Vryl · · Score: 3, Insightful
      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' ?

    4. Re:The game is Slashdot, the score is Karma. by Anonymous Coward · · Score: 0

      It wouldn't be easy, even for chess. You could get the "correct" rules, but you could get a statistical version which fitted perfectly (eg 20% of pawns are lost in the first N moves)[1]. Castling rules just by watching - forget it. [1] Like quantum physics?

    5. Re:The game is Slashdot, the score is Karma. by dlkf · · Score: 1
      This is not about learning the rules, this is about learning to score the most points. To do this, you will usually have to learn the rules. If this was about learning the rules only, the programs would be asked what all the legal moves are at each time step, not asked to pick the best move from a given list.

      How can one consider this contest artificial intelligence?

      The idea is that by learning how to play the games, the programs will become more intelligent. And the one that learns the fastest and thus scores the most points will be the most intelligent. This contest is designed mostly for people in the field of machine learning which is usually considered a subset of the field of AI.

  61. Re:random fortune... by Dunedain · · Score: 1

    Yes. It starts out with preconceptions, but you don't know what they are. The "right way" to do it is to have uniform weights, so you *know* that it has no preconceptions... but that means using a quite different learning algorithm. It's the difference between random and representative. If you flip a coin ten times, HHHHHHHHHH is just as common as HTHTHTHTHT, but only one is representative.

    --
    -- Brian T. Sniffen
  62. But the real question is by Anonymous Coward · · Score: 3, Funny

    Google, raging, or lycos?

  63. Paranoia by Anonymous Coward · · Score: 1, Interesting

    Don't you think this could be used for piloting weapons ? The site is dead at this time so I can't get more insight. yours, Britney

  64. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    I guess I don't agree with your viewpoint because history doesn't concur with your view of it. It was a good lighthearted joke though... thanks for making me laugh.

  65. Re:Offtopic: Trip to Israel (now totally offtopic) by Betcour · · Score: 1

    Well the Palestinian have a clue about the whole thing. The Israelian can't win, as they are not facing an organised army but a population where every kid over 7 is a potential enemy. For Israel to win, they'd have to kill every Palestinian.

    The Palestinian have nothing to loose, most of them are rather happy to die in this war, and the Israelian have everything to loose.

  66. Re:This should prove entertaining. by ichimunki · · Score: 2

    I thought it was called "presidential election".

    --
    I do not have a signature
  67. This should prove entertaining. by BiggestPOS · · Score: 3, Interesting
    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.

    --
    What, me worry?
    1. Re:This should prove entertaining. by Anonymous Coward · · Score: 0

      :) moreso than we can know...

    2. Re:This should prove entertaining. by tardibear · · Score: 1, Insightful

      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 ..

    3. Re:This should prove entertaining. by Anonymous Coward · · Score: 0

      +5 Insightful.

    4. Re:This should prove entertaining. by rabidcow · · Score: 1

      There's a card game called "Mao".

      You get a penalty for asking about the rules.

      I probably shouldn't even tell you that much...

  68. That recall me a couple of clever hack... by Anonymous Coward · · Score: 5, Interesting
    ...where the problem was to make a program that played the old paper/rock/scissor game.

    The entries had to be given in the form of a subroutine that played the next move (given the current score and the history). The judges were linking two of them together and run the resulting binary.

    Of course, there have been an entry that looked in the stack and modified the scores.

    But the greatest was one (IMHO) that fork()ed and returned one possible response in each of its child. At next turn, the one that did not make the point (ie: had top score), exit()ed.

    Mind-blowing. Found the link

    That program was the "Fork Bot"

    Cheers,

    --fred

    1. Re:That recall me a couple of clever hack... by Anonymous Coward · · Score: 0

      Which goes to show why you don't leave the real AI research to the peanut gallery. i.e. you all.

  69. Aha! by spiral · · Score: 1

    > ...wait until the player showed up there, blowing him away upon entry into the game.

    So that's where all those camping bastards came from!

    --
    Drinking will help us plan!
  70. Is there a page for this? by Perianwyr+Stormcrow · · Score: 2

    Sounds like an amusing thing to watch while it grows...

    --

    What we call folk wisdom is often no more than a kind of expedient stupidity.-Edward Abbey

  71. AI? by Anonymous Coward · · Score: 1, Funny

    Give me a woman's brain anyday!
    Sincerely, Mike Bouma

  72. Re:Two thoughts by Black+Parrot · · Score: 1

    > If I win, do I get a trip to the US? (see email address)

    Furriners who win a trip to the US shouldn't come unless they also get a get out of jail free card.

    --
    Sheesh, evil *and* a jerk. -- Jade
  73. Re:Even if I could write this program i wouldn't.. by drc500free · · Score: 1

    Oh, I don't know... It's more like Taco is renting from Joe, who told him that he would be selling him the apartment. Meanwhile Joe promises the apartment to his cousin Frank, evicts Taco and gives it to Frank. Now Taco keeps coming back and leaving piles of burning shit on Frank's doorstep, and trying to sleep in his kitchen. At which point Frank starts beating him with a baseball bat. Neither side can claim that they are being mature about this, but the problem is that Joe is an ass, and was never held responsible for tooling them both around...

  74. Why not pick a real problem? by janpod66 · · Score: 1, Flamebait

    There are enough real, interesting problems out there to choose from; why pick some company's idea of a contest? Work on Go, write a nice chess player, do something interesting with data mining, etc.

    1. Re:Why not pick a real problem? by Quixote · · Score: 2, Insightful

      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".

    2. Re:Why not pick a real problem? by DPalomo · · Score: 1

      As stated in the article, the submission of the winner will be used to improve natural language processing. I think this is **quite** a real world challenge.

      Daniel.

      --

      - For every winner, there are dozens of losers. Odds are you're one of them -

    3. Re:Why not pick a real problem? by janpod66 · · Score: 2

      Well, there are lots of interesting, open, well-defined, important problems in natural language processing. Why not work on them directly?

  75. Re:I might know how to win or get an unfair advant by bprotas · · Score: 2, Insightful

    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...

  76. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    I think you completely missed the point of what I wrote above. There was a time when all 3 people of the major faiths lived together, why is there a problem now?

    Let me give you a quick "timeline" of the problem and show you why your post is just a rehash of what I read in the papers.

    1. Palestinians call the land of current day israel home (more than a century ago) --> 2. The land where Israel is taken and the native palestinians forced out of their land/homes (close to 50 years ago) --> 3. Misc Battles and skirmishes --> 4. Israeli attacks --> 5. palestinians attack/defend

    You're post would make sense if the timeline started at #5. Unfortunately, you can't blame the palestinians for their desire to have their homes back. The timeline starts at #1, and quite frankly (and the UN of most of the nations of this earth agree) that Israel provoked the attacks when they savagely killed and took away the land of the palestinians.

    Answer this, if today the an outside force came into Tel Aviv and threw out every Israeli public official and told the locals to move out, wouldn't the Israeli's have a field day with it? Wouldn't they cry injustice? Wouldn't they use their weapons against the "barbarians" trying to take their homes? Isn't it wrong of the "invaders" to take their land? If you are being honest with yourself, you must answer yes to all those questions. But I ask you something else, aren't we being hypocritical when we didn't ask those same questions when the Israeli's took over with their tanks and weapons so long ago?

    How do you justify state sponsored assassinations by the israeli's? Do you remember the boy and father that were brutally murdered by the Israeli army? (will post link later if you really want to see it)

    To sum it up, yes the palestinians may be doing stuff that is wrong, but the Israeli are no angels. The Palestinians show signs of a people who have nothing to lose. Look at their children that are alive. They hurl stones at tanks. Do you think that is a normal child? Doesn't that bring tears to your eyes? Children are very amazing in that they see the world differently. But one thing they do understand is if something upsets their parents. These children do this on their own accord--if that doesn't give you a clue I don't know what will. Their homes taken, children killed, spouses murdered, jobs lost, and lives turned upside down by the same people (israeli's) that cried for help during hitler's time.

    I don't mean to sound harsh, but I honestly don't understand how you can even attempt to defend the Israeli's.

  77. I can beat tit-for-tat by CyberDruid · · Score: 1
    Beating the tit-for-tat is easy when you are allowed to submit many entries. Submit a lot of fake progs that defect against everyone who does not reply to a specific ID string (transmitted by the early moves).

    The real prog answers an ID string with something similar and voíla the fake entries becomes nice cooperators. (see my also my earlier post)

    --

    Opinions stated are mine and do not reflect those of the Illuminati

  78. random fortune... by Pelam · · Score: 5, Insightful
    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.
    1. Re:random fortune... by jesser · · Score: 2, Informative

      I don't get it. Is Minsky trying to say that using a random neural net doesn't mean that the net won't have any preconceptions?

      --
      The shareholder is always right.
  79. All Sample games only work by Collusion! by rufusdufus · · Score: 1

    They give 3 samples, rock scissors paper , prisoners dilemma, and cooperation. The only way to gain advantage above random in Rock Scissor Paper is to collude with the other player for wins. The cooperation game gives points if the two players pick numbers sufficiently close to one another--again the only win vs random is to collude with the other player. Finally, the prisoners Dilemma has the odd property that the best score comes from colluding with the enemy. Can this be a coincidence? What are they aiming for here? See how many people are smart enough to write a bunch of programs that collude with themselves?

  80. Re:Even if I could write this program i wouldn't.. by drc500free · · Score: 1

    okay, wasn't a perfect example, but the real problem IS that the british promised Palestine to both the Palestinians and the Jews. So though the Jews were given the land, the Palestinians have a right to be pissed.

  81. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    Jewish Suffering Deserves a Better Legacy
    Copyright: http://www.iviews.com
    Published Saturday January 06, 2001

    By Fahhim Abdulhadi
    Israel has stated one primary reason for rejecting the Palestinian refugee's right to return to their homes: it would threaten Israel's character as a Jewish state. This assertion has a familiar ring. Blacks were seen as a threat to the 'white' character of apartheid-era South Africa. Appropriately, the South African attitude met with international revulsion. It is time to reject the same sort of state sponsored racism from Israel.

    Palestinians represent the largest refugee group in the world, some 4 million people according to the UN. This is equivalent to the population of Colorado. Their plight began in 1948, when some 750,000 Palestinians were forced to flee before advancing Israeli forces. Zionist leaders, including Israeli founder-father David Ben Gurion, saw forcible transfer of the land's Palestinian population as a legitimate solution to the "Arab problem." Israeli historian Benny Morris has reported that in no instance did an Arab population leave their homes before an attack by Jewish forces.

    On December 11, 1948 the United Nations passed Resolution 194, a document with a very specific and just solution to the refugee problem. It says "refugees wishing to return to their homes and live at peace with their neighbors should be permitted to do so at the earliest practicable date, and that compensation should be paid for the property of those choosing not to return." Israeli agreement to abide by this stillunimplemented resolution was a precondition for Israel's 1949 admission into the UN.

    For decades the refugees have waited on justice. While they have waited, many have lived and died in filthy refugee camps. While the children of Palestine's refugees play next to and in human excrement, Israel has become the largest single recipient of US foreign aid. As a result of this huge investment, the fruit of a well-organized Washington lobby, Israelis have one of the best standards of living in the Middle East. Refugees on the other hand, are still frequently worrying about where to get their next meal. I toured refugee camps in Gaza and Jordan in 1999.

    In Gaza's Beach Camp I met a family with ten children where the recently crippled father could not walk or earn a living. An aid worker told me that the family received a UN package each month that fed them for less than a week. After that they were dependent on their neighbor's good will. Of course, I, as a guest, was invited to stay for dinner. I declined and still feel shamed by such genuine hospitality.

    The idea that people forced to flee their homes and raise their children in such horrible conditions will be denied the dignity of return to their homes is horrific. That the reason for this denial might be that they had the misfortune of being born as non-Jews should stop any sane person in their tracks. Khalil Shikaki, a Palestinian political analyst, rightly points out, "(The refugees) have lived for 50 years like this. If you tell them now there is no right of return, it's like spitting in their faces." Be sure of one regrettable reality, a signed deal will not stop the refugees resentment at being turned into a bargaining chip from boiling over. Sacrificing their right to justice is sacrificing a lasting peace.

    Zionists, followers of a political movement first articulated in the late 1800's that called for the establishment of a Jewish national home on any available territory, are acutely aware that their nation was built boot on face of another people. Given that the injustice of Jewish suffering is the primary reason presented for the need to create a state, such action seems rather hypocritical.

    The tragedies of the Jewish people do not justify the Zionist transfer of that suffering to Palestinians, whose only crime is to be non-Jews. Certainly, Zionist concerns over the ethnic purity of Israel can in no way be taken as a serious reason to deny Palestinians a right to return to the land where they were born. Jewish suffering deserves a better legacy.

  82. hmm by JM_the_Great · · Score: 1

    Sounds like it's time for a genetic algorithm plus a bit of matrix game theory.

    --

    --Justin Mitchell
    "2nd Place is a fancy word for losing" --Bender (Futurama)
  83. Are you that arrogant? by Anonymous Coward · · Score: 0

    The Israeli's can't "kick" anyone's ass. They are cowards that hide behind their weapons. If the Israeli's were that good, they'd exterminate every palestinian out there. They can't though. The world already knows and believes that the Israeli's are at fault. And no religion preaches anything about purposefully killing another person. Why do you try to pin the blame on Islam? What you are seeing is a fight behind the desire to live and see the sun rise again the next day.

    Its people like you that are so blind to the real situation that you let the inhumanity continue. Do some research before talking out of your ***.

  84. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    Could you share your sources with everyone? I'm really interested in hearing about this so called UN resolution.

    The IDF targets innocent human beings. Ambulance workers, children, unarmed civilians. Its a war that Israel started when they conqured the land. I think there is another post below that asks how the country of Israel was created. I think you should answer that question. Was it barren land and no one lived there? If you say yes, then you are a liar and history will tell otherwise. History shows very clearly David Ben-Gurion had a lot to do with the murderous campaign conducted against the native palestinians. Please do not attempt to refute this because only you will look like a fool.

  85. Trip ... by fogof · · Score: 0, Flamebait

    You mean trip to palestine... right ?

    --
    --=.=-- www.cyber2000.qc.ca
  86. Re:Even if I could write this program i wouldn't.. by drc500free · · Score: 1

    "For the palestinians, there could be peace with the israelis alive. Sure, I bet some palestinians would be harassed, even if they are innocent of crimes (though I'm not sure there are any like that). Yet the palestinians never hesitate to take action that leads to their primary goal, which is the genocide of the israelis and other jewish people. Doesn't that say something to you?"
    Making generalities about people is exactly the problem here... I have met plenty of Palestinians who don't want to kill the Israelis, and plenty of Israelis who don't want to kill the Palestinians.
    Okay - I'm done. No more political crap from me ;-)

  87. Damn by tsa · · Score: 1

    Pity I can not program computers. This looks like a nice challenge.

    --

    -- Cheers!

  88. Quite frankly, Sweetz0r by theneo · · Score: 0, Insightful

    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.

  89. How many actual AI researchers reading slashdot? by exa · · Score: 4, Funny

    I mean how many real world CS people studying AI reads Slashdot?

    Doing that and dumb enough to waste his time with this. Count me one.

    --
    --exa--
  90. tit-for-tat algorithm by jesterzog · · Score: 4, Interesting

    If this is a 2+ player competition and they're the right sorts of games (like the rock-paper-scissors game that it mentioned), whoever wins it might have to figure out a way to consistently beat the tit-for-tat algorithm.

    Tit-for-tat is one of the dead simplest game playing algorithms, and collectively it's one of the most successful.

    It's based on the rule of "always do what the other player did last move". Under most circumstances it's impossible for it to actually win a game because the other player is always one step ahead. But its strength is in winning tournaments.

    While it always loses, it never loses by much. This is different from other algorithms which usually have about as many weaknesses as they have strengths and will usually flunk out in at least some trials.

    If someone can beat it consistently in a tournament situation, they really will have accomplished something in AI. Of course, this whole thing depends on exactly how the rules are structured, the scoring system and the information available to the program.

    1. Re:tit-for-tat algorithm by Anonymous Coward · · Score: 0

      In the case of Rock-Paper-Scissors, you don't see the enemy move before yours, so I assume you mean the program will do what the enemy did last round.

      It also won't work if you get more/less turns than your enemy.

      If the enemy recognizes your behaviour and understands a game like rock/paper/scissors, you're toast.

    2. Re:tit-for-tat algorithm by jareds · · Score: 2

      What are you talking about? Tit-for-tat wins prisoner's dilemma tournaments, not games in general. The name is even based on prisoner's dilemma, it doesn't make sense to call repeating your opponents's last move tit-for-tat in general.

      Also, you could trivially beat it at rock-paper-scissors: just play rock, paper, scissors, rock, paper, scissors, and so on forever. You will win every round except maybe the first one. Finally, if you actually read the rules of this competition, you would find that there's no guarantee that there will always be one opponent move for every one of your moves, so you couldn't even implement tit-for-tat.

  91. A lot easier for AI when... by fractaltiger · · Score: 1

    If only God open sourced the brain...

    --
    "Wireless : LAN :: Laptop : Desktop"
  92. Two thoughts by Cave+Dweller · · Score: 4, Insightful

    1. If I win, do I get a trip to the US? (see email address)
    2. I wonder whether the winner could visit me.

    :-)

  93. Exercise in neural networks by slasho81 · · Score: 2, Insightful

    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.

    1. Re:Exercise in neural networks by exa · · Score: 1

      I don't think NN will work out. Why do people here think that neural nets are any good? NNs are slow things. You can use the most advanced NN only to solve moderate problems with low quality and enormously slowly.

      This is a general machine learning task, and NN is not the only method. As a matter of fact in most of machine learning data mining literature NNs are not seen as state-of-the-art. There are much more advanced algorithms.

      --
      --exa--
  94. oh my a trip to israel by Anonymous Coward · · Score: 0

    i hear the fireworks are nice

  95. I might know how to win or get an unfair advantage by CyberDruid · · Score: 4, Interesting

    Seems to me that since it is a round-robin for all contestants (the site was /.ed, but I saw another post claiming that this was the case), all you have to do is team up with a lot of friends and have them enter fake programs into the contest (i.e cheating). These programs will start by identifying themselves with an "ID-string", consisting of, say, the first 10 replies (this can obviously be done generically even with unkown rules, just pick the moves randomly with the same seed). When my program sees this ID it replies a similar code. When the fake programs sees this, they start cooperating with my program (by playing as badly as they can muster). If the fake programs does not get this reply, they start playing as well as they can and will (since there will probably be large element of luck in each game) steal a considerable amount of points from the pool. The "real" program never risks anything since it never sends its own ID before being statistically sure that the opponent indeed is a fake. This method was inspired by a similar trick in the famous Prisoner's dilemma game.

    --

    Opinions stated are mine and do not reflect those of the Illuminati

  96. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    right on brother.

  97. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    Please dont talk about something you dont know shit about.

  98. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0, Troll

    Yeh. Those evil Israaeli's, picking on poor little arabs whose religion (cult?) just so happens to demand that they utterly destroy Israel and jews in general. I mean, why not instead invite the palestinians over for milk and cookies? The israelis have never made it priority 1 to genocide palestinians, and there is evidence that suggests they'd rather be doing other things. Hmm. Do you think that maybe its the palestinians, that provoke all this?

    Think about it. The israelis, when they go after some palestinian, or take action, it's always after some crime has been committed. They go after those that they have a reasonable suspicion of causing it (most of the time, but there are abuses). They don't go out and shoot 10 random palestinians. On the other hand, what else can you call it, when the palestinians blow up a crowded street? They surely can't hope to take out specific targets, so they can't be going after specific enemies. They are more or less randomly killing israelies, regardless of any percieved crime or guilt.

    For the israelis, there will never be peace until all palestinians are dead, and yet they hesitate to kill them all. Does anyone doubt they have the military might to do so? And if they don't, doesn't that attest to their general decency?

    For the palestinians, there could be peace with the israelis alive. Sure, I bet some palestinians would be harassed, even if they are innocent of crimes (though I'm not sure there are any like that). Yet the palestinians never hesitate to take action that leads to their primary goal, which is the genocide of the israelis and other jewish people. Doesn't that say something to you?

    And one other thing. Don't you think it's funny that Arafat gives speeches to arab audiences, where he points out places in the Koran that say its ok to pretend peace with your enemy, as long as it gives you time to build an army to destroy them? That's not even subtle folks.

  99. rand() by KurdtX · · Score: 2, Interesting

    I took an AI class this year where we had a challenge to use PERL to design a Stratego-playing AI. One of the professors quickly wrote a script that moved a random piece a random direction (verifying the move was legal), and had a surprisingly high win %.

    --

    Kurdt
    I'm not anti-social. Just pro-technology.
  100. humans? by aozilla · · Score: 1

    Hopefully they'll throw in some human opponents, just for comparison sake.

    --
    ok then your [sic] infringing on my copyright! Could you as [sic] me next time before STEALING my comments for your own?
  101. How they pay for the prizes... by A+nonymous+Coward · · Score: 5, Insightful

    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!

    1. Re:How they pay for the prizes... by Tablizer · · Score: 0
      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.

      It probably has happened, but only a fool would tell the world how.

    2. Re:How they pay for the prizes... by mysta · · Score: 1
      They are playing the STOCK MARKET.

      Wow! Day-trading must have gone to new extremes then. Check out one of the conditions of entry:

      Your program must be able to make at least ten moves per second on the tournament machine.
      --

      "Where is the wisdom we have lost in knowledge, and where is the knowledge we have lost in information?"-T.S.Eliot
  102. Virii by sui · · Score: 0

    "In fact, you are free to submit a binary executable if you desire: there is no need for you to reveal your source code to us! This interface allows an external judge program to moderate head-to-head games between two programs easily." Hope they got a good virus scanner

    --
    Why do the kids in West Side Story have to join a street gang if they can afford $70 Gap khakis?
  103. 95% of the solution is written already: by wickline · · Score: 1
    1. Re:95% of the solution is written already: by tyoud1 · · Score: 1

      That's really cool; do you know where the source can be found? I've looked around on google, hoping to find it, but I've had no luck.

  104. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    The israelis are hardly guilty of invading the land, as the original poster suggested.

    Israel did invade the land. That they do so isn't disputed by anyone with even the slightest knowledge of the circumstances. Life is complicated and there certainly are those that believe that the Israelis had some cause to invade, whether historical (or biblical) claims to the land or in the interest of being in a better defensive position themselves. That they invaded, however, is not in doubt.

  105. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    Last I checked, the Israeli government is elected by the people, thus the majority of them DO hate palestinians enough to kill them. FWIW, the palestinians suck too.

  106. Re:Glossary by rueba · · Score: 1

    Cool! I actually learned something on /. ....I am just really getting down with this field of AI especially Machine Learning, but I like what I am learning, hopefully I will be able to make a contribution......

    Rob

    --
    The only reason all cover-ups appear to fail is that you never hear about the ones that succeed.
  107. Glossary by rueba · · Score: 2
    As an aspiring(NOTE:aspiring) AI guru, I decided to "googlize" some of the cryptic abbreviations.

    Backprop - The backpropagation algorithm for training neural networks. "Brought neural networks back from the dead."-popular opinion

    RBF - Radial Basis Function Neural Nets.

    HMM - Hidden Markov Models(I don't know what they're for)

    SVM - Support Vector Machine(latest hot thing,I still don't know why its so great)

    SOFM - Self Organizing Feature Maps(another NN architecure)

    ART - no idea here... Like Hal Cohen's Aaron?(the drawing program)

    I would like to add that I think Computational Learning Theory(COLT) and other statistical approaches seem really promising and I think we are going to be seeing some really interesting stuff in the next 5 - 10 years(ok maybe 20).

    --
    The only reason all cover-ups appear to fail is that you never hear about the ones that succeed.
    1. Re:Glossary by Anonymous Coward · · Score: 1

      ART is Adaptive Resonance Theory, brought to you by Steve Grossberg, Gail Carpenter, et al. ART is actually a bunch of different architectures. It has both interestingly useful and disturbingly undesirable properties that are a little too intricate to get into here.

      SVMs are cool since unlike the others, they're built from the ground up for classification problems - using a RBF with least squares computation will be optimal if your error function is the squared error, but in the case of discrete classes, its not. Applying RBFs & Perceptrons to discrete classification is kind of a hack that people frequently overlook, but they generally work well anyway. If you have the computer power & the problem is worth the effort of testing another method (e.g. stock market) for unknown results, go for it. SVMs are uncool since their computation time can get into the hours on an Athlon 1Ghz, while RBF & Perceptron can finish in 1 sec. SVMs are also a PITA to code relative to the others. BTW the new round of SVM research was triggered by the Sequential Minimal Optimization algorithm by John Platt of Microsoft Research - parts of MS really are innovating.

      Hidden Markov Models (HMMs) are used in time series problems: speech recognition/classification/production, stock market, some vision.

      I sincerely wish you well with your AI efforts, don't take the comment you responded to as a tirade against newbies, just against people that haven't learned that it takes years to study an academic field, and that its always much deeper than you'd imagine. I think its safe to say that noone without a Phd is going to contribute significantly to creating the first strong AI, but there's a ton of cool stuff that can be built with modern techniques, if only the general programming populous knew about them. The problem is that the profs usually see devising applications as beneath them; they tend to throw their network at some benchmark datasets, publish & never really apply the same algorithm again, which I think is a mistake.

      Some good texts: "Neural Networks for Pattern Recognition" by Christopher Bishop, "Pattern Recognition and Neural Networks" by Brian Ripley, and "Pattern Classification, 2nd ed" by Duda, Hart & Stork (or the original "Pattern Classification and Scene Analysis" by Duda & Hart).

  108. God used perl? by Tablizer · · Score: 0

    Re:"A lot easier for AI if only God open sourced the brain... "

    He/she *has*. The problem is that it is trillions of line of source code that is harder to read than Perl and takes a lot of biology to unravel.

  109. prize by Anonymous Coward · · Score: 0

    that's an old, old joke. first prize - a one week trip to Israel. second prize - a two week trip to israel. third prize - ...

  110. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    A pity they missed you.

  111. There is a whole subfield of AI about learning by exa · · Score: 1

    Naive bayesian is one of the weaker methods.

    Thanks,

    --
    --exa--
  112. Re:Even if I could write this program i wouldn't.. by Anonymous Coward · · Score: 0

    Your raving sounds as if Israel is an Evil Empire filled with Evil Scientists doing nothing but Evil Military Research.

    The truth is that despite a very difficult security situation Israel is a major contributor of new ideas and technologies used by the civillian population worldwide. Instant messaging has originated in Israel, and so did the Intel MMX technology.

  113. Silence is golden by exa · · Score: 1

    You are so very right. Got to keep my mouth tighter.

    --
    --exa--
  114. Re:Even if I could write this program i wouldn't.. by Josh+Kramer · · Score: 1

    I am going to guess that you are currently living in Israel. Thank you for providing an insiders look into what is really going on over there. It is a shame that people take the word of CNN, BBC, AP, etc.. as the ultimate truth. To those that would bash Israel for Genocide, or indiscriminate killing, or war mongering, perhaps you should learn some history: did you know that over 800,000 Jews were expelled from Arab lands when Israel was created (Where are the calls for their compensation), lookup such things as Black September, the massacre of the Lebanese Christians in Damur, the closing of the Straits of Tiran (Closing an international waterway is an Act of War), the PLO charter is good read, as is the book From Israel to Damascus(which should shut up anyone claiming Sharon is responsible for Sabra and Chatilla, check out chapter 8). If you don't like history try current events such as the hypocricy of the world leaders at the G8 (condemn Israel for firing on Armed terrorists but yet using live ammo on unarmed protestors in Italy is OK), or that the UN finally admiting that it misled Israel about the kidnappings by Hizbollah (not only did they have one tape but there are three!!! and personal belongings such as blood stained uniforms (of which they will NOT return), no bias here just keep on moving), perhaps learning about the Palestinian media would be enlightening (calls of hate and violence are not uncommon). If sheeple persist is continuing their ingnorance there can never be peace because there will be no understanding.

  115. The reason for such a generic interface... by lisrael · · Score: 1

    Comaparing the main area of research for the company sponsering this competition, and the rules of the competition, I suspect they are going to try to use these bots to make some progress on the Turing test. They defined 'game' widely enough to consider conversation a game. The score is the reward/punishment - relevent and/or sensible responses are rewarded, etc. I suspect they want to give the machines thousands of possible moves, each representing a phrase, and then set them loose 'against' each other, seeing which one can score the highest on the conversation scale.

  116. Re:How many actual AI researchers reading slashdot by Reality+Master+101 · · Score: 2

    AI hasn't progressed in 50 years. Its a failure.

    I'm one of those "chumps" that would say something like this. Quite frankly, "Artificial Intelligence" is an abject failure.

    Just because there isn't "strong AI" yet doesn't mean the field has failed.

    Ah, there we are. If you don't want your field to be called an "abject failure", then don't call it AI. If you want to say that pattern recognition has made some strides, say it. Or pick your particular problem. But when you use the words "artificial intelligence", then I expect intelligence created artificially. Not "very sophisticated algorithms".

    Physics doesn't have a grand unified theory, medicine can't make people live forever, etc. & we don't consider those fields failures.

    No. But comparing an AI scientist to a physicist is like comparing Aristotle to Einstein. Both brilliant people, both made huge contributions to their field, but immense differences in knowledge.

    I'm not saying AI will never get there, but I think a dose of reality is needed here.

    --
    Sometimes it's best to just let stupid people be stupid.