Forgot your password?
typodupeerror
Programming Games

545-Person Programming War Declares a Winner 57

Posted by Soulskill
from the bring-me-the-severed-subroutine-of-your-fallen-foe dept.
An anonymous reader writes: A while back we discussed Code Combat, a multiplayer game that lets players program their way to victory. They recently launched a tournament called Greed, where coders had to write algorithms for competitively collecting coins. 545 programmers participated, submitting over 126,000 lines of code, which resulted in 390 billion statements being executed on a 673-core supercomputer. The winner, going by the name of "Wizard Dude," won 363 matches, tied 14, and lost none! He explains his strategy: "My coin-collecting algorithm uses a novel forces-based mechanism to control movement. Each coin on the map applies an attractive force on collectors (peasants/peons) proportional to its value over distance squared. Allied collectors and the arena edges apply a repulsive force, pushing other collectors away. The sum of these forces produces a vector indicating the direction in which the collector should move this turn. The result is that: 1) collectors naturally move towards clusters of coins that give the greatest overall payoff, 2) collectors spread out evenly to cover territory. Additionally, the value of each coin is scaled depending on its distance from the nearest enemy collector, weighting in favor of coins with an almost even distance. This encourages collectors not to chase lost coins, but to deprive the enemy of contested coins first and leave safer coins for later."
This discussion has been archived. No new comments can be posted.

545-Person Programming War Declares a Winner

Comments Filter:
  • I thought Greed was "The Multimillion Dollar Challenge" where teams of five tried to answer trivia questions but each round one player was randomly paid to try to take another player out of the game, or be thrown out trying....

  • by Anonymous Coward on Friday June 13, 2014 @03:23PM (#47232225)

    and upload this to the drone control network.

  • The final ranking was performed with a 673-core computer cluster, which simulated 153,439 games in under one hour for just $5.74
    • They sure have! We just spun up 20 c3.8xlarge spot instances on Amazon, AWS is pretty magical.
      • by lgw (121541)

        Spot especially, as it's relatively cheap - quite cheap as supercomputing cluster time goes!

    • Please, I ran a simulation just like this on a 1992 super computer, and I haven't paid a cent.

      I guess the agreement did say payment was due when processing finishes next decade, though.

  • by upontheturtlesback (2605689) on Friday June 13, 2014 @03:30PM (#47232287)
    This is really interesting and exciting work. In 2010, we showed that nearly this exact algorithm is used by neonates (newborns) to govern their visual attention and eye movements, and it explains much of what we know about newborn visual attention. It's exciting to see that when you essentially parallelize the algorithm with multiple agents that are aware of each other, it becomes an extremely efficient algorithm for resource collection in a completely different field/task. http://www.ncbi.nlm.nih.gov/pu... [nih.gov]
    • by macklin01 (760841) on Friday June 13, 2014 @04:25PM (#47232641) Homepage

      We also use very similar force algorithms in our cancer models. :-) e.g., http://www.sciencedirect.com/s... [sciencedirect.com]

      The description of the agents and forces in this summary was actually very well done.

    • by macklin01 (760841)
      BTW, that's a very cool paper! Have you considered using these techniques towards image processing, segmentation, etc.?
      • That's really cool! I find it really interesting and elegant to see the same simple model describe the behavior of such disparate systems that, on the surface, look complicated, but can be described by the sum of simple mechanisms.

        I agree, the summary was really well written.

        That's a good question, about using similar techniques for image processing and object segmentation from a scene. From a cognitive standpoint, neonates rapidly build on this simple model over their first few months of life as th
    • by John Bokma (834313) on Friday June 13, 2014 @04:37PM (#47232749) Homepage
      "Free full text" but no link to download the PDF (at least, I don't see it). PDF: http://psych.mcmaster.ca/maure... [mcmaster.ca]
      • by gringer (252588)

        You need to click on the "Elsevier Open Access" link from NCBI, which is a direct link to the article on the publisher's website (this location is where you click for all PubMed articles, as long as the publisher has provided access in that way). PubMed never displays complete articles.

        After clicking through, there's a "Download PDF" link at the top left of the article, just under the green Science Direct header.

        • Thanks, for the clear explanation. It's extremely confusing to put it nicely.
          • by gringer (252588)

            It's extremely confusing to put it nicely.

            I feel compelled to tell the world about a more confusing part of NCBI that I'm trying to navigate myself around at the moment: The Transcriptome Shotgun Assembly Sequence Database. Submitting sequences is... a little tricky. Here's a simplification of the process:

            1. Create a BioSample record for the organism that you're submitting data for
            2. Download a sample template tab-delimited file, and fill in arbitrary descriptions about your organism
            3. Upload the template file (usin
    • by smaddox (928261)

      Anyone else find it odd that he used a distance squared force for a 2D problem? The surface of a circle depends linearly on the radius.

      • by cbhacking (979169)

        The goal is that the attraction to the coins is greater when you're close (so you don't wander past one, pulled by that large cluster off to the side) and the repulsion is lesser when you're further away (so two allies can turn directly towards each other to pick up coins that lie between both, even though they repel each other).

      • Anyone else find it odd that he used a distance squared force for a 2D problem? The surface of a circle depends linearly on the radius.

        Linearly being the key word... take it one step at a time (before looking at what geometry inverse square law could represent). The rule is derived entirely from distance... Distance reduces the number of spacial dimensions into one, it doesn't matter how many spacial dimensions you have so long as you can find a scalar distance between two points.

        For a less abstract explanation think of a 2D simulation as a geometrical subset of a 3D simulation (that subset doesn't have to be axis aligned), a 2D simulation

    • It actually sounds like a "Schema Architecture" that Arkins proposed in 1998 http://mitpress.mit.edu/catego... [mit.edu]. You can implement it in about 10 lines of python because it's just that: the sum of attractive (goals) and repelling (obstacles) force vectors, weighted by the inverse of the distance squared. I was surprised OP didn't mention the Schema architecture, because it is exactly that, and since it sounds like a (simulated) robot game...

      Your paper on newborn looking is really interesting. I build robotic

  • Core War (Score:5, Informative)

    by sremick (91371) on Friday June 13, 2014 @03:44PM (#47232365)
  • by Charliemopps (1157495) on Friday June 13, 2014 @04:17PM (#47232583)

    This is a very fun game! I've been looking for stuff like this. Normally I have fun writing stuff like this in games until they ban me for "Hacking" when really the hacking was the only fun part of me. Now the hacking bit IS the game.

    Maybe I can ditch my EVE mining bots now ;-)

  • Freakin' coders. (Score:4, Interesting)

    by Kazoo the Clown (644526) on Friday June 13, 2014 @06:14PM (#47233367)
    You can tell these guys are all lamer coders, they can't document worth squat. In the forum some guy asks for clear docs and they repond in essense with "just run our simulator, it's too complicated to explain." What a bunch of hosers. A competition like this ought to have clearly deliniated parameters. From reading their page I can't tell a darn thing about what the "Greed" environment is, what the problem to be solved is, and the summary of the winning solution on the Slashdot article here presumes you already know exactly what the conditions and goal with which the warring program must run. I see references on the linked contest site to coins that "randomly appear" and not much more. There's no way he could submit his solution to a journal except the "Journal for Irreproducible Results." Lazy bastards. There may be an interesting solution to something here, but there's seems no way to tell exactly what without reverse engineering their simulator.

After an instrument has been assembled, extra components will be found on the bench.

Working...