I'm not sure why more people aren't more suspicious of this story.

The guy is traveling overseas. A $500,000 investment property of his was sold by supposed Nigerian scammers. He rushes home, after being notified by his neighbors, to halt the sale of a second investment property.

And, if you believe the comments about Australian law, he will get compensated for the sale of the house while the buyers get to keep it.. and this $500,000 that was wired to China just disappears.

Since the owner stands to benefit the most from this scam, I'm inclined to believe he is the scammer. Not some dude in an internet cafe in Abuja, Nigeria.

Mainly.. because a lot of commenters here are assholes, and it seems fashionable to dismiss anything that isn't 80% certain to happen.

Its not hard to imagine the entire internet becoming a sort of gargantuan parallel processing swarm intelligence (or the infrastructure being utilized to create an AI based on Swarm Intelligence).

There's some quote that I cannot even remotely remember, but can be paraphrased as such:

People assume that their limitations are the limitations of man.

Anyway.. humanity advances (at times) due to the exceptional works of less than 0.001% (yeah, you heard me.. the population times 0.00001) of the people that are living. (No citations here, just lazy speculation).

I suppose the rate of technological advancement could just sort of fizzle out over the next 20 years.. with the only new inventions being things like cheaper toasters and more fuel efficient vehicles and bigger televisions and faster computers.. but I sure hope the future holds more than that.

I concur! The concept of particle swarm optimizations is so simple.. (and me being fairly dumb, me like so much).

One particular approach I like is where you have randomly defined neighborhoods for your particles. And as the simulation progresses.. members of the neighborhood that do well get more random connections to other algorithms while ones that do worse lose connections.

Then, the particles just move towards (with some random perturbations) whatever member of the neighborhood is doing "better" (however you wish to define that) at any given time.

So... your particles swarm over your search space.. and form into these different neighborhood clumps around potential solutions.. and also discover new solutions while moving towards an old one.

For me, its more intuitive than some of the other genetic algorithm approaches where you have algorithms dying off and creating offspring. Instead they're like the flying monkeys in the Wizard of Oz, "Fly, my pretties! Fly! Bring me my solution!"

Well.. if you read their pdf (linked at the bottom of the blog post), you see that they literally turn the Fitness Function into a trainer that leads teams to the most proper ways to play (see pg5 Section "4.5 Fitness evaluation").

The first step in the learning process is that the teams should spread out on the field and be relatively evenly distributed.

Second was, move players closer to the ball.

Third, kicking the ball was given value.

Fourth, getting the ball closer to the opponents goal.

Then finally, the most weight was given to scoring goals.

So.. they have a great system here.. but it is mainly suggesting that these fitness guidelines eventually lead to behaviours that we understand as playing good defense or whatever.

To be fair, I'm guessing that maybe Mr. Elmenreich was using the word "trainer" in some literal sense.. They didn't need some trainer for other parts of the system since that was built into the fitness evaluation step.

What I want to know is: Did they put in a penalty for offsides?

Well, it's even a little less amazing than you suggest.

If you read the pdf that the blog post links to ("Evolving neural network controllers for a team of self-organizing robots"), on Page 5 in section "4.5 Fitness Function"... they discuss how they decompose fitness down from simply "Score the most goals" to little component tasks that "ensure a smooth learning process assuming some preliminary knowledge or ideas about the solution."

They practically lead the algorithms to better solutions along certain paths. The first thing they want the algorithms to learn is that "a good distribution on the field might lead to good overall play." So, now you have a "trainer" telling the players on a team to spread out.

So.. while their framework looks interesting.. and reading the pdf is good food for thought and helps me think about my own problem domains.. they seem to have restricted the system to prevent any truly odd (yet successful) behaviours from evolving. Thus.. their fitness function encourages successful behaviours that they expect to see.

"Same as it would be gross to smoke a turkey inside or any number of things..."

Somehow... that just feels like a lead-in to a good joke..

:) good thing someone knows what I'm saying without everything needing to be spelled out.

For those with the need for more info, find "Back Door Man" by Howlin' Wolf.

For the record, I would have no problem being a back-door man until I turn 40. Little girls understand they need their back-door man.