The current phase of the contest will end December 18th at 11:59pm EST. At that time submissions will be closed. Shortly thereafter the final tournament will be started. The length of the final tournament has not yet been determined but is expected to last less than one week. Upon completion the contest winner will be announced and all results will be publically available.
Gogo LessWrong team! The experience and the potential publicity will be excellent.
I’ll chip in with a prize to the amount of ($1000 / team’s rank in the final contest), donated to the party of your choice. Team must be identified as “LessWrong” or suchlike to be eligible.
This sounds like a wonderful opportunity for anyone interested to promote Lesswrong and themselves as well as give to a good cause like SIAI/other worthy charity! We should really bring this to people’s attention.
It also sounds like an excellent test of applied rationality.
It seems like there’s a decent amount of interest. This should probably be made into a post of its own, and hopefully a promoted one, if we want an official LessWrong team. A lot of people don’t check back on the open thread who would probably be interested in joining.
This should probably be made into a post of its own, and hopefully a promoted one
Yes I think it should be, I think there would be some interest. Considering we have some very competent and experienced people on Lesswrong and some very enthusiastic amateurs, several teams wouldn’t be too bad an idea either if there where enough people. Some of the amateur LWers might be a bit intimidated by being part of the “Offical LessWrong team”, whereas “Lesswrong Team #3″ or “Lesswrong amateur rationalist group” dosen’t sound as bad.
I’d consider joining a team thing. A LessWrong team would be cool if it, you know, wins… Currently there is not much tough competition, my bot is incredibly stupid (doesn’t pay attention to the other players) and in the top 200.
I know about it from the contest they held last year.
I’d be interested in joining a team—I’m a video game programmer with an AI degree, so it’s the kind I should be good at (I don’t have massive amounts of free time though).
My most important thought was to ensure that all CPU time is used. That means continuing to expand the search space in the time after your move has been submitted but before the next turn’s state is received. Branches that are inconsistent with your opponent’s move can be pruned once you know it.
Architecturally, several different levels of planning are necessary: food harvesting and anticipating new food spawns. Pathfinding, with good route caching so you don’t spend all your CPU here. Combat instances, evaluating a small region of the map with alpha/beta pruning and some pre-tuned heuristics. High level strategy, allocating ants between food operations, harassment, and hive destruction.
If you’re really hardcore, a scheduling algorithm to dynamically prioritize the above calculations. I was just going to let the runtime handle that and hope for the best, though.
Is anyone writing a bot for this contest?
http://aichallenge.org/index.php
Sounds awesome, where did you first hear of this?
Anyone interested in starting a team for this?
Gogo LessWrong team! The experience and the potential publicity will be excellent.
I’ll chip in with a prize to the amount of ($1000 / team’s rank in the final contest), donated to the party of your choice. Team must be identified as “LessWrong” or suchlike to be eligible.
This sounds like a wonderful opportunity for anyone interested to promote Lesswrong and themselves as well as give to a good cause like SIAI/other worthy charity! We should really bring this to people’s attention.
It also sounds like an excellent test of applied rationality.
It seems like there’s a decent amount of interest. This should probably be made into a post of its own, and hopefully a promoted one, if we want an official LessWrong team. A lot of people don’t check back on the open thread who would probably be interested in joining.
Yes I think it should be, I think there would be some interest. Considering we have some very competent and experienced people on Lesswrong and some very enthusiastic amateurs, several teams wouldn’t be too bad an idea either if there where enough people. Some of the amateur LWers might be a bit intimidated by being part of the “Offical LessWrong team”, whereas “Lesswrong Team #3″ or “Lesswrong amateur rationalist group” dosen’t sound as bad.
Done: http://lesswrong.com/r/discussion/lw/8ay/ai_challenge_ants/
I’d consider joining a team thing. A LessWrong team would be cool if it, you know, wins… Currently there is not much tough competition, my bot is incredibly stupid (doesn’t pay attention to the other players) and in the top 200.
I know about it from the contest they held last year.
Is this an annual event?
Seems to be. This is their third contest.
I’d be interested in joining a team—I’m a video game programmer with an AI degree, so it’s the kind I should be good at (I don’t have massive amounts of free time though).
I put some thought into it, but I don’t think I’ll have time to. I wouldn’t mind sharing my ideas with anyone who is actually doing it.
I wouldn’t mind listening to ideas...
My most important thought was to ensure that all CPU time is used. That means continuing to expand the search space in the time after your move has been submitted but before the next turn’s state is received. Branches that are inconsistent with your opponent’s move can be pruned once you know it.
Architecturally, several different levels of planning are necessary: food harvesting and anticipating new food spawns. Pathfinding, with good route caching so you don’t spend all your CPU here. Combat instances, evaluating a small region of the map with alpha/beta pruning and some pre-tuned heuristics. High level strategy, allocating ants between food operations, harassment, and hive destruction.
If you’re really hardcore, a scheduling algorithm to dynamically prioritize the above calculations. I was just going to let the runtime handle that and hope for the best, though.