AI Advances

Nate Truitt writes "Research published in the journal "Artificial Life" reveals that autonomous agents -- software entities capable of independent action in dynamic, unpredictable environments -- can communicate with one another as they work to achieve a common goal. They actually develop a language of sorts...." It doesn't look like the paper is available yet, but I know that Giles has been willing to send copies by email of his other papers.

Comments

communicating in their own language? hmmm....

[Tumbleweed]

Why do I get this uneasy feeling, and remember "Colossus" and "Guardian" from the movie "The Forbin Project"? Yikes...

like ants???

[killthiskid]

Hmm... so agents in 2-d were able to communicate to locate another 'prey' agent.

Could this be applied to something like gnutella, where agents live on the n-dimesional grid of the gnutella world, with each server having hosting a 'message board' where agents that know about one another communicate? The prey would be a specific search, be it keyword or filename.

Since these agents were supposedly able to create a language that used words with contextual meanings, would that allow for search in-context?

Ok, perhaps a bit of a stretch, but it does make sense that searches of distrubeted networks are going to have to use some sort of searching agent.

Ants use something like this. When many ants are being succesful at a certain task, more ants do that task... i.e. if food is plentiful, ants coming back from foraging indicate to other atns that their search was succesful. After enough messages like this, an ant will shift gears.