|
|
|
Dik-Dik Van Dik
Skeptic Friend
United Kingdom
76 Posts |
 Posted - 05/14/2005 : 12:20:27
|
Anyone interested in artificial life? I've always been fascinated by AL ever since I played the game 'creatures', maybe its because I want to play God or something, but it was really fascinating to breed those little things and they were really quite well designed, they had genes (which could mutate during breeding), body chemicals, neural nets etc.
For most kids i guess the fun was to treat these things as pets and perhaps breed a funny colour creature. but I didnt really care how they looked and just left them to run wild and let natural selection run its course. And I wasn't the only one. Special breeding programs were created by net enthusiasts which actually 'exported' weak creatures from the habitat at certain intervals. Soon though thhis became dull as the program was glitchy and despite lots of fun variations, no true speciation could ever occur.
Having just finished a BSc in AI though I'm finding it pretty hard to find any jobs relating to Artifical intelligence and NONE relating to artificial life (i'm guessing work in the field is almost totally research). Even if i were to get into AL as a hobby I can't find anything on the net that would allow me to experiment with it. Short of writing my own programs and stuff, which to be honest is pretty beyond me.
Are a few kids games really it when it comes to artificial life in your own home?
|
DARWIN 3:16
"The simple believeth every word." - Proverbs 14:15
|
Edited by - Dik-Dik Van Dik on 05/14/2005 12:26:59
|
|
|
Dr. Mabuse
Septic Fiend
Sweden
9687 Posts |
Posted - 05/14/2005 : 15:19:31 [Permalink]
|
Many games employ AI to guide opponents, so my first thought is that developing AI for game programming companies would be a possible career. I daydream sometimes about learning how AI and neural networking is made. But my time is too short, there are so many other projects... I wish I was independently wealthy, so I could spend all my time doing cool things instead of slaving at Ericsson Mobile Communications for my salary.
Developing control-programs with AI for exploring drones and similar for Moon and Mars would be cool. Or perhaps for controlling prosthetics. A guy I know lost both his arms in an accident, one at the shoulder, the other one just above the elbow. He would benefit from a prosthetic smart enough to "guess" what he want it to do.
Prosthetics have evolved much the last decades, thanks to new materials and robotics.
(Edit: grammar or something) |
Dr. Mabuse - "When the going gets tough, the tough get Duct-tape..."
Dr. Mabuse whisper.mp3
"Equivocation is not just a job, for a creationist it's a way of life..." Dr. Mabuse
Support American Troops in Iraq:
Send them unarmed civilians for target practice..
Collateralmurder. |
| Edited by - Dr. Mabuse on 05/14/2005 15:56:26 |
 |
|
|
Dik-Dik Van Dik
Skeptic Friend
United Kingdom
76 Posts |
Posted - 05/14/2005 : 15:37:44 [Permalink]
|
True that artificial intelligence has many uses and applications. But artificial LIFE seems like one of those branches of research that is fascinating but not lucrative..
If you're really interested in neural nets, its really not that complex, i did my dissertation/project on voice-print I.D using neural nets. Managed to get about 92% accuracy using a simple perceptron.
If you wanna mess with them, i'd use a program like matlab. Its got a NN toolbox and lots of nice built in things so you don't have to write a function for every little thing you wanna do. |
DARWIN 3:16
"The simple believeth every word." - Proverbs 14:15
|
| Edited by - Dik-Dik Van Dik on 05/14/2005 15:39:34 |
|
|
|
Plyss
Skeptic Friend
Netherlands
231 Posts |
Posted - 05/15/2005 : 12:46:43 [Permalink]
|
quote:
Originally posted by Dik-Dik Van Dik
True that artificial intelligence has many uses and applications. But artificial LIFE seems like one of those branches of research that is fascinating but not lucrative..
If you're really interested in neural nets, its really not that complex, i did my dissertation/project on voice-print I.D using neural nets. Managed to get about 92% accuracy using a simple perceptron.
If you wanna mess with them, i'd use a program like matlab. Its got a NN toolbox and lots of nice built in things so you don't have to write a function for every little thing you wanna do.
Artificial life doesn't seem lucrative at the moment, indeed. I do foresee a time when it may become so, but that's probably a long time in the future. Nevertheless, cool stuff like this (.wmv) fills me with hope (movie from this webpage).
I've never played "creatures" but it sounds interesting. I've always found the field of artificial life fascinating, but i never really pursued that interest. All i know about the subject comes from a popular book by Chris Langton (which i highly recommend to anyone interested in the subject).
I've given some thought on a few projects i could try for myself, even with my minimal programming experience, but i just have to find the time to do it. Also, i really don't know computers or mathematics, so that hardly helps as well. Ah, if only there were 48 hours in each day :( |
|
|
|
latsot
Skeptic Friend
United Kingdom
70 Posts |
Posted - 06/04/2005 : 01:21:39 [Permalink]
|
quote:
Even if i were to get into AL as a hobby I can't find anything on the net that would allow me to experiment with it. Short of writing my own programs and stuff, which to be honest is pretty beyond me.
There are quite a few alife resources about the place with software to download. For example:
http://www.cs.ucl.ac.uk/staff/t.quick/alife.html
http://www.alife.net/
http://black.hollygo7.co.uk/dysturbwiki/index.php/Software
However, I would encourage you to develop your programming skills if you want to play with alife. You really just need to build your confidence through practice. There are plenty of books and web resources out there to get you started. I have a few books covering artificial life left over from when I did some work in the field about a decade ago. When I get chance, I'll dig out some references.
As for jobs, you are probably right that they are mostly research-based. There certainly is some alife stuff going on in games development - for example, in squad-based games (how do you make the squad move realistically?) and in large-scale battle-based games (how do teams form, split, reform etc?)
For a similar reason, there is work in alife going on in the military for training purposes and simulation.
Alife has also been proposed for a number of more real-world applications such as in telecoms (optimisation etc.)
The major downside of alife is similar to that of neural nets. Would you trust a system that works because of emergent behaviour that you can't analyse? Who can you blame if your system does something stupid or dangerous? How would you do due diligence on a system like this?
The problem is more of a legal one than a technical one - those of us working in software know better than to trust *any software whatsoever* anyway. But it is partly incidental practical considerations like this that have prevented wider adoption of AI techniques in mainstream applications.
Hope this helps
r |
|
|
|
Siberia
SFN Addict
Brazil
2322 Posts |
Posted - 06/04/2005 : 09:26:58 [Permalink]
|
Artificial intelligence/life is one of the reasons why I'm on college, majoring in Computer Science... actually, it IS the only reason.
All I can think, though, for some research that is also fun, is game development. Sure, AI has lots of uses, but for ALife, I'd think games. I loved the game Creatures, though I played it but briefly |
"Why are you afraid of something you're not even sure exists?"
- The Kovenant, Via Negativa
"People who don't like their beliefs being laughed at shouldn't have such funny beliefs."
-- unknown
|
|
|
|
Ricky
SFN Die Hard
USA
4907 Posts |
|
|
dv82matt
SFN Regular
760 Posts |
Posted - 06/04/2005 : 10:46:13 [Permalink]
|
Gene Pool is an facinating A-Life program which demonstrates the effects of sexual selection.
Here's the main page of the site. It's good stuff. |
|
|
|
latsot
Skeptic Friend
United Kingdom
70 Posts |
Posted - 06/06/2005 : 08:05:31 [Permalink]
|
quote:
Originally posted by dv82matt
Gene Pool is an facinating A-Life program which demonstrates the effects of sexual selection.
Here's the main page of the site. It's good stuff.
Can you simulate the effects of sexual selection? Im not sure this program does. Interesting, though.
Cheers
r |
|
|
|
latsot
Skeptic Friend
United Kingdom
70 Posts |
Posted - 06/08/2005 : 11:48:30 [Permalink]
|
Hey Ricky - completely off-topic but good to run into you on another forum. Hi.
r |
|
|
|
dv82matt
SFN Regular
760 Posts |
Posted - 06/08/2005 : 21:32:26 [Permalink]
|
quote:
Originally posted by latsot
Can you simulate the effects of sexual selection? Im not sure this program does. Interesting, though.
I think that it does simulate the effects of sexual selection (or maybe just something similar to it. There is only one sex after all.), but I should mention that I was actually more impressed with Version 3 of Gene Pool in that it starts out with randomly assembled organisms (as opposed to Version 4 which starts out with predesigned organisms which then appear to de-evolve) which then evolve based on fitness (the ability to efficiently get food bits and chase after mates) and attractiveness (the ability to attract a mate.)
Anyway if you're interested I'd recomend giving Version 3 a whirl. |
|
|
|
Baxter
Skeptic Friend
USA
131 Posts |
Posted - 06/08/2005 : 22:42:07 [Permalink]
|
| I remember 'Creatures.' I thought I was the only person who actually had that. Thanks for the nostalgia... |
"We tend to scoff at the beliefs of the ancients. But we can't scoff at them personally, to their faces, and this is what annoys me." ~from Deep Thoughts by Jack Handey
"We can be as honest as we are ignorant. If we are, when asked what is beyond the horizon of the known, we must say that we do not know." ~Robert G. Ingersoll
|
|
|
|
latsot
Skeptic Friend
United Kingdom
70 Posts |
Posted - 06/15/2005 : 06:50:26 [Permalink]
|
quote:
I think that it does simulate the effects of sexual selection (or maybe just something similar to it. There is only one sex after all.)
How could you have sexual selection with only one sex?
quote:
but I should mention that I was actually more impressed with Version 3 of Gene Pool in that it starts out with randomly assembled organisms (as opposed to Version 4 which starts out with predesigned organisms which then appear to de-evolve) which then evolve based on fitness (the ability to efficiently get food bits and chase after mates) and attractiveness (the ability to attract a mate.)
Im a bit worried about the word 'de-evolve'...
Cheers
r |
|
|
|
Dude
SFN Die Hard
USA
6891 Posts |
Posted - 06/15/2005 : 11:16:26 [Permalink]
|
quote:
How could you have sexual selection with only one sex?
Different sexes are not required for meiosis to occur. Sexual reproduction, in its simplest form, is nothing more than the joining of two hapliod nuclei (formed by meiosis) into a diploid nuclei.
The sexual reproductive phase of yeast, and other Ascomycota, are an example of this.
|
Ignorance is preferable to error; and he is less remote from the truth who believes nothing, than he who believes what is wrong.
-- Thomas Jefferson
"god :: the last refuge of a man with no answers and no argument." - G. Carlin
Hope, n.
The handmaiden of desperation; the opiate of despair; the illegible signpost on the road to perdition. ~~ da filth |
|
|
|
|
latsot
Skeptic Friend
United Kingdom
70 Posts |
Posted - 06/18/2005 : 00:56:18 [Permalink]
|
quote:
Originally posted by Dude
quote:
How could you have sexual selection with only one sex?
Different sexes are not required for meiosis to occur. Sexual reproduction, in its simplest form, is nothing more than the joining of two hapliod nuclei (formed by meiosis) into a diploid nuclei.
The sexual reproductive phase of yeast, and other Ascomycota, are an example of this.
Yes....but can you explain how sexual *selection* could occur with only one sex?
Sexual selection happens when traits in one sex of a species are (for some reason - lets not open that can of worms for the moment) desirable to the other sex. These traits are selected because the opposite sex finds them desirable in a mate: they will therefore tend to choose mates with those traits, resulting in the propagation of those genes throughout the gene pool
The most-cited example is the peacock's tail. It doesn't seem to have any particular purpose other than to be attractive to peahens. If for some reason peahens happen to like big tails with eye-patterns, they will be more likely to mate with the individuals that possess this trait - a clear case of selection.
There is a lot more to the theory of course - we have to ask at some point why peahens happen to like peacocks with fancy tails and we have to consider the evolutionary arms races that can occur between the species (the genes causing peahens to prefer fancy tails will also propagate, for example). In fact, sexual selection asks (and provides some answers for) some of the deepest questions in evolutionary science.
Darwin considered sexual selection to be exceedingly important (he wrote a book about it) and it is often known as "Darwin's other theory". Sexual selection arguments have been convincingly proposed to explain all kinds of things - why humans have little body hair, the evolution of human intelligence and many others.
I'm just having trouble understanding how sexual selection could occur without sexes.
Cheers
r |
|
|
|
Dave W.
Info Junkie
USA
26020 Posts |
Posted - 06/18/2005 : 04:26:10 [Permalink]
|
quote:
Originally posted by latsot
I'm just having trouble understanding how sexual selection could occur without sexes.
Easy: you have a trait which makes you "hot" to all the other individuals. If a trait makes individuals prefer to mate with only those carrying that trait, it doesn't matter if there's only one sex, two sexes or 17 sexes, it'll be selected for based on mating habits, and perhaps nothing else. |
- Dave W. (Private Msg, EMail)
Evidently, I rock!
Why not question something for a change?
Visit Dave's Psoriasis Info, too. |
|
|
 |
|
Topic Locked
