Jump to content

  • Log In with Google      Sign In   
  • Create Account

Interested in a FREE copy of HTML5 game maker Construct 2?

We'll be giving away three Personal Edition licences in next Tuesday's GDNet Direct email newsletter!

Sign up from the right-hand sidebar on our homepage and read Tuesday's newsletter for details!


We're also offering banner ads on our site from just $5! 1. Details HERE. 2. GDNet+ Subscriptions HERE. 3. Ad upload HERE.


Emergent Intelligence


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
73 replies to this topic

#61 BrianL   Members   -  Reputation: 530

Like
0Likes
Like

Posted 23 August 2004 - 04:42 AM

All of this started over the initial comment:

"Emergent behavior" is really just a fancy way of saying "a coincidence that really looked cool"

I think everyone here agrees that this is a simplistic exaggeration (intentionally, to illustrate a point). The primary problem people saw with this statement is that it doesn't require 'rules' (psuedo randomness isn't emergence) or a reoccuring sense-act cycle (the result of the application of physics an isn't emergent behavior).

From an outside observers stand point, InnocuousFox and Timkin appear to be arguing roughly the same position, just using different terms to describe it.

Can we let this thread die peacefully?

Sponsor:

#62 IADaveMark   Moderators   -  Reputation: 2509

Like
0Likes
Like

Posted 23 August 2004 - 05:22 AM

Quote:
Original post by eeaiguy
See my articles in AI Game Programming Wisdom...
Yeah... no kidding. I forgot about that. From "AI Game Programming Wisdom", Article 1.2 ("The Illusion of Intelligence"), page 19:
Quote:
(begins with quote from Dyson that I had seen earlier in my Google search but didn't post)
Emergent behavior is that which cannot be predicted through analysis at any level simpler than that of the system as a whole... Emergent behavior, by definition, is what's left after everything else has been explained. [Dyson97]

We can use emergent behavior (EB) to give the illusion of intelligence to a game AI. In fact, in many cases, occurances of EB an even provide the illusion to the developers themselves! At one point, during the testing of Empire Earth, one computer player attempted to expand to an area on the other side of an enemy's town. As his citizens were attempting to break through, a massive attack force came in and kept the enemy busy while the citizens got through. This was not programmed in - the attack just happened to coincide with the expansion attempt due to the adjustment of timers governing both behaviors. [IF: italics and bold mine]

Unfortunately, it is very hard to purposely create EB; rather, one needs to create an architecture in which EB can occur...


I find it strikingly "coincidental" that Bob used the word "coincide" in his description above.

Quote:
we've actually talked quite a bit about EB *as it pertains to games*. And he's right about it in that context.

I have no doubt that Timkin is right about it in the grander scheme of AI, but that doesn't normally apply to those of us writing AI for games.
My point exactly... initially and throughout. In fact, I repeated that contextual point in my later post where I said:
Quote:
Well, I'm just repeating what game AI designers, programmers and litterature consider "emergent behavior" to be. Perhaps it is different in the "real world"... but seeing as this is a game AI board, I didn't bother addressing contexts other than that of game AI.

(Back to quoting eeaiguy)
Quote:
Not that some of the tools of academia aren't used by us or that we don't appreciate the work being done by them...
And, contrary to what Timkin was alleging, I agree with this.

Good to see you, Bob. Thanks for checking in.

#63 IADaveMark   Moderators   -  Reputation: 2509

Like
0Likes
Like

Posted 23 August 2004 - 05:36 AM

Quote:
Original post by RPGeezus
Please delete this thread-- it is a stain on the forum.
Careful. A few months ago I deleted an entire thread that I had started and in which Timkin had nutted out on me for something unrelated. I figured the thread, as is, was a "stain on the forum." He chewed my ass out royally saying I had no right to delete (my own) thread. He even reported me to Dave Astle (The Great Sightless Rhino of the West). I talked to Dave about it at GDC... who frankly thought the whole thing was rediculous.

While I agree with you that this thread went beyond stupid at some point, my point is that if this thread is deleted, there may be some sort of spontaneous fission reaction Down Under.
Dave Mark - President and Lead Designer of Intrinsic Algorithm LLC

Professional consultant on game AI, mathematical modeling, simulation modeling
Co-advisor of the GDC AI Summit
Co-founder of the AI Game Programmers Guild
Author of the book, Behavioral Mathematics for Game AI

Blogs I write:
IA News - What's happening at IA | IA on AI - AI news and notes | Post-Play'em - Observations on AI of games I play

"Reducing the world to mathematical equations!"

#64 IADaveMark   Moderators   -  Reputation: 2509

Like
0Likes
Like

Posted 23 August 2004 - 05:41 AM

Quote:
Original post by BrianL
All of this started over the initial comment:

"Emergent behavior" is really just a fancy way of saying "a coincidence that really looked cool"

I think everyone here agrees that this is a simplistic exaggeration (intentionally, to illustrate a point).
Oh my God! Someone actually GETS the idea of oversimplifying to illustrate a point! I thought that if people didn't realize what I was doing the first time, they certainly would get it later when I acknowledged I was being "flipant and terse" and may have "overcompensated". Thank you for acknowledging the fact that I was possibly not expecting my words to be taken literally.

Quote:
Can we let this thread die peacefully?
Shit, I was trying to let the whole thing die back on page 1 when I said
Quote:
"However, since we have determined that this is not the goal of the OP, it is not appropriate for me to detail the differences here."
Aparently someone didn't want to let it go at that.

[Edited by - InnocuousFox on August 23, 2004 12:41:14 PM]

#65 RPGeezus   Members   -  Reputation: 216

Like
0Likes
Like

Posted 23 August 2004 - 06:09 AM

ENOUGH ALREADY!!

Quote:

Careful. A few months ago I deleted an entire thread that I had started and in which Timkin had nutted out on me for something unrelated. I figured the thread, as is, was a "stain on the forum." He chewed my ass out royally saying I had no right to delete (my own) thread. He even reported me to Dave Astle (The Great Sightless Rhino of the West). I talked to Dave about it at GDC... who frankly thought the whole thing was rediculous.

While I agree with you that this thread went beyond stupid at some point, my point is that if this thread is deleted, there may be some sort of spontaneous fission reaction Down Under.


How was that meant to help anything?

Dave, If Tim is making you angry, ignore him or send him a private message.

Tim, if Dave is making you angry, ignore him or send him a private message.

This is NOT the lounge.. You don't see this stuff happening in the graphics forum. The Math and Physics forum gets along nicel.. So what's wrong with the AI forum?

I'm going to ask Timkin to kill this thread as he is the moderator here. If he refuses, I will ask someone else to kill the thread.

Will

(P.S. I've giving both InnocousFox and Timikin a negative rating for this one. Not for the initial disagreement, but for the way in which it degenerated. You're both long-time members and should know better).



[Edited by - RPGeezus on August 23, 2004 2:09:20 PM]

#66 IADaveMark   Moderators   -  Reputation: 2509

Like
0Likes
Like

Posted 23 August 2004 - 06:55 AM

I just go an email from Ferretman. He's back from being gone over the weekend. I was the one who initially pointed out this thread to him anyway last week and we have been discussing it. I figure he will close it when he pops in... and I would welcome it.
Dave Mark - President and Lead Designer of Intrinsic Algorithm LLC

Professional consultant on game AI, mathematical modeling, simulation modeling
Co-advisor of the GDC AI Summit
Co-founder of the AI Game Programmers Guild
Author of the book, Behavioral Mathematics for Game AI

Blogs I write:
IA News - What's happening at IA | IA on AI - AI news and notes | Post-Play'em - Observations on AI of games I play

"Reducing the world to mathematical equations!"

#67 Anonymous Poster_Anonymous Poster_*   Guests   -  Reputation:

0Likes

Posted 23 August 2004 - 08:20 AM

Quote:
We can use emergent behavior (EB) to give the illusion of intelligence to a game AI. In fact, in many cases, occurances of EB an even provide the illusion to the developers themselves! At one point, during the testing of Empire Earth, one computer player attempted to expand to an area on the other side of an enemy's town. As his citizens were attempting to break through, a massive attack force came in and kept the enemy busy while the citizens got through. This was not programmed in - the attack just happened to coincide with the expansion attempt due to the adjustment of timers governing both behaviors.


uhh...that actually just sounds like a coincidence, not emergent behavior. if a fedex guy shows up just as a plane crashes into your bedroom, saving your life, it doesnt suggest any kind of system at work. we may be getting "flippant" with our definitions, but lets not go willy nilly.

#68 IADaveMark   Moderators   -  Reputation: 2509

Like
0Likes
Like

Posted 23 August 2004 - 10:57 AM

Quote:
Original post by Anonymous Poster
uhh...that actually just sounds like a coincidence, not emergent behavior. if a fedex guy shows up just as a plane crashes into your bedroom, saving your life, it doesnt suggest any kind of system at work. we may be getting "flippant" with our definitions, but lets not go willy nilly.
Remember that' Bob's definition was included in the chapter regarding the "illusion of intelligence". In this sense there is a fine line between coincidence and "unintended interaction".
Dave Mark - President and Lead Designer of Intrinsic Algorithm LLC

Professional consultant on game AI, mathematical modeling, simulation modeling
Co-advisor of the GDC AI Summit
Co-founder of the AI Game Programmers Guild
Author of the book, Behavioral Mathematics for Game AI

Blogs I write:
IA News - What's happening at IA | IA on AI - AI news and notes | Post-Play'em - Observations on AI of games I play

"Reducing the world to mathematical equations!"

#69 BenevolentLiao   Members   -  Reputation: 144

Like
0Likes
Like

Posted 23 August 2004 - 12:42 PM

Success is not the only drop scene of a research project. Take a review of the history of Prolog, at 1990 age it is the hottest research field of AI, but now where is it? A researcher can just only got respect by himself. Academic EB research can not offer any useful tool to AI domain currently, they does not need any apology by anyone. All academic EB demonstration seems very idiotic, it is why AI programmer disdain them.

Quote:
Original post by Ferretman
......THIS IS THE WORSE KIND OF EMERGENT BEHAVIOR THAT SEEMS TO BREED IN FORUMS....



#70 Anonymous Poster_Anonymous Poster_*   Guests   -  Reputation:

0Likes

Posted 28 August 2004 - 02:52 PM

One of the main ideas in Swarm Inteligence (at least what I got from it as being new) was the use of patterns of localized solution sets being tested to converge faster on an acceptable solution (and to avoid local maxima/minima problems).

The parameters of a random pattern is adjusted semi-inteligently
matching the problem type and the progress being made in the test cycles.

Its the equivalent of using a Monte Carlo method except making a educated guess as to what subset of the solution space to apply it to.

#71 Anonymous Poster_Anonymous Poster_*   Guests   -  Reputation:

0Likes

Posted 28 August 2004 - 02:55 PM

Quote:
Original post by haphazardlynamed
nice
your armies learned that world peace was the best answer
takes a bunch of triangles to figure that out what people cant...



Throw uncertainty (on several levels) into it and your Triangles are powerless.

#72 Nathaniel Hammen   Members   -  Reputation: 136

Like
0Likes
Like

Posted 29 August 2004 - 06:37 AM

Quote:
Original post by Samith
Here's an example of sort of emergent behavior I got once. I made a program that pitted 2 armies of triangles against eachother. The rules were, if you killed a guy on your own army you lost a some points, and if you killed a guy on the other army you gained some points. Every few minutes, I stopped the battle, and the guys with the best score moved on and "breeded" a new army. Then I would start over. Well, what I expected was that these armies would fight, but that's not what happened. At the beginning of each battle, the triangles would always shoot all of their bullets right away and never move. Obviously, since my triangles were lined up in rows and columns, a lot of friendly fire was happening. Well, after a few generations, my armies learned to shoot all of their ammo right away but not kill eachother. Right when the battle started they would all turn to a precise angle such that they could shoot everything and not kill anything. It was pretty neat, even though they weren't doing exactly what I had wanted them to do.


Just curious, but were the units blind, or could they "see" where their enemies were? Also, were you using a GA or a GP?

I am the master of ideas.....If only I could write them down...

#73 Samith   Members   -  Reputation: 2260

Like
0Likes
Like

Posted 29 August 2004 - 02:03 PM

The triangles could "see" the nearest enemy and the nearest friendly triangle. I don't know the difference between GA and GP, so I'll explain to you how I did it: Each triangle had a neural network to make decisions for it, and I used to weights in the NN as genes in the genetic processes. I would mix and match them and randomly mutate some of them after every round.

#74 Nathaniel Hammen   Members   -  Reputation: 136

Like
0Likes
Like

Posted 29 August 2004 - 03:53 PM

The weights as the genes...
I guess that would be considered a Genetic Algorithm... I think.

Anyways, I plan to make something like this, except I will use a Genetic Program to decide which enemy that the units can see. It will probably not come up with the "peacful solution" that your units came up with, because my units will probably not choose to see enemies that have an obstruction in the way. Or they will choose not to shoot if they do see it... probably.

Edit: too...many...typos!

[Edited by - Nathaniel Hammen on August 29, 2004 10:53:35 PM]

I am the master of ideas.....If only I could write them down...




Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS