Jump to content

  • Log In with Google      Sign In   
  • Create Account


no-one can create ai


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
95 replies to this topic

#41 MikeD   Members   -  Reputation: 158

Like
Likes
Like

Posted 10 July 2003 - 10:17 PM

quote:
Original post by Anonymous Poster
That has an easy explanation, people are used to see cars and drive them, but people aren''t used to deal with snakes or spiders. That''s definately not ''hardcoded'' in our brains.


People grow up seeing cars every day. People also grow up seeing spiders almost every day (unless you''re living in some oxygen tent there is probably a spider within a metre of you right now). Arachnophobia is several orders of magnitude bigger than a phobia of cars. You do the math.



Sponsor:

#42 UlfLivoff   Members   -  Reputation: 122

Like
Likes
Like

Posted 10 July 2003 - 10:25 PM

Ok - let''s take another example. Place a hungry newborn baby on the mother and it will automatically crawl to the breasts. Now how did it know that there was gonna be food there ??

This are mere examples and not explanations, but If you still disagree, I suggest that you read some books on the topic. I''m not referring to a specific book, anyone will do

Ulf

#43 Pipo DeClown   Members   -  Reputation: 804

Like
Likes
Like

Posted 10 July 2003 - 10:28 PM

No one can create Intelligence, that''s why we have Artificial Intelligence.

.lick


#44 MikeD   Members   -  Reputation: 158

Like
Likes
Like

Posted 10 July 2003 - 11:40 PM

Pipo DeClown: You haven''t really got the hang of reasoned arguments have you

Mike

P.S. The same could occasionally be said of many here including myself

#45 ahw   Members   -  Reputation: 262

Like
Likes
Like

Posted 11 July 2003 - 04:05 AM

ulflivoff : the smell of milk my friend. I suggest you smell a few titties, and you''ll see the truth in that answer
The smell triggers reactions in the baby, like opening the mouth, grasping, etc. Note how a baby, when offered anything that vaguely has the shape of a nipple, will suck without ever questioning what it is that it is sucking.

Who said the whole suckling mechanism was a proof of intelligence ? If you wanted to show it was a hardcoded behaviour, good news, it is. Otherwise I am not sure where you are getting at.


Sancte Isidore ora pro nobis !

#46 UlfLivoff   Members   -  Reputation: 122

Like
Likes
Like

Posted 11 July 2003 - 04:47 AM

quote:
Otherwise I am not sure where you are getting at


If you read the previous posts then it''s pretty obvious what we''re discussing.

It''s funny how some programmers with no experience in psychology are 100% confident in their own homemade psychological theories.

At least I''ve read a few books on the topic and it''s their theories am referring to here...

Reminds me of Bertrand Russels wise words:

The problem with humans is, that stupid people are always 100% confident in what thei''re doing is right, where as intelligent people area always doubtful.

#47 Timkin   Members   -  Reputation: 864

Like
Likes
Like

Posted 12 July 2003 - 03:48 PM

quote:
Original post by UlfLivoff
Ok - let''s take another example. Place a hungry newborn baby on the mother and it will automatically crawl to the breasts. Now how did it know that there was gonna be food there ??



Newborn babies don''t crawl...

There are several hard-coded stimulus-response behaviours that babies have that enable them to take a nipple and an autonomic suckling action as well. We''ve evolved with these... but put a baby out of reach of a lactating breast and it won''t know where to go or how to get there... it might smell the milk and get excited... but that''s a different story all together!

Timkin




#48 Code-Junkie   Members   -  Reputation: 122

Like
Likes
Like

Posted 12 July 2003 - 04:05 PM

"Newborn babies don''t crawl..."

My thoughts exactly, Timkin.

#49 UlfLivoff   Members   -  Reputation: 122

Like
Likes
Like

Posted 12 July 2003 - 09:20 PM

Ups, My bad

he he

[edited by - UlfLivoff on July 13, 2003 4:29:16 AM]

#50 MikeD   Members   -  Reputation: 158

Like
Likes
Like

Posted 13 July 2003 - 12:54 AM

Newborn babies of _our_ species don''t crawl.

But apparently the word baby can mean other species as well (I looked it up to be sure).

Take the Kangaroo and other marsupials for example. The baby Kangaroo, looking like a foetus and being absolutely tiny in proportion to a Joey, (which is a young Kangaroo in case you didn''t know) crawls up the mother''s fur on birth and finds it''s way into the pouch for the last several months of development into a young Kangaroo, attaching itself to the mother''s teat inside the pouch. I can''t say it''s performing lactotaxis but it performs a set of hard-coded behaviours involving crawling towards milk the second it''s born. These Kangaroo babies are approximately 2.5 centimetres (1 inch) in length.
Some argue that the reason human''s are so unable to care for themselves at birth is because the high level of plasticity and potential for adaptation that we have necessitates a lack of hard coding at birth (we still have hard coding but a lot less than (perhaps almost) all other species).
The detachment from hard wiring in the brain might allow our bodies and brains to evolve more swiftly (you evolve the body but can''t evolve it away from the hard coding, so you can''t evolve it very far before letting the hardcoding catch up). This brings into question whether evolvability itself is an evolutionary advantage giving an individual increased fitness on an evolutionary scale. It probably is.

Mike

#51 UlfLivoff   Members   -  Reputation: 122

Like
Likes
Like

Posted 13 July 2003 - 07:59 AM

Phief!! Saved by MikeD again! Just like the time in the politics forum.

I owe you one Mike

LOL

#52 Timkin   Members   -  Reputation: 864

Like
Likes
Like

Posted 13 July 2003 - 05:11 PM

quote:
Original post by MikeD
Newborn babies of _our_ species don''t crawl.

But apparently the word baby can mean other species as well (I looked it up to be sure).



Yeah, okay, point taken! Since we were suggesting that animals can be intelligent, it''s fair to consider baby animals too!

quote:

Some argue that the reason human''s are so unable to care for themselves at birth is because the high level of plasticity and potential for adaptation that we have necessitates a lack of hard coding at birth (we still have hard coding but a lot less than (perhaps almost) all other species).



From my understanding of the common literature (i.e., not academic literature) on babies, it is suggested that we are so helpless because we are born far earlier than we should. Gestation is thought to be 9 months because after that time period, the babies head has grown too large to fit through the small hole in the female pelvis, which is roughly 10-12cm in diameter. One could speculate that very early in our evolution we could remain in the womb longer (since we would have had smaller heads) and would have been more capable of looking after ourselves (with parental assistance) than we are now at birth.

One could then postulate an evolutionary advantage of adaptability as our brain size increased. I.e., consider two mutations of the species, both with increased brain size, but one with higher adaptability and one with more hard-wiring. Both would need to be born earlier than optimal gestation due to the increased head diameter. One could speculate that the more adaptable mutation would be more likely to survive in successive generations because the parents would be adaptable to dealing with a helpless baby, as opposed to the hard-wired parents, who would do the same old routine. Clearly I''m over-simplifying the issue, but hopefully you catch my drift!



quote:
Original post by MikeD
This brings into question whether evolvability itself is an evolutionary advantage giving an individual increased fitness on an evolutionary scale. It probably is.



I think it''s fairly obvious actually that adaptation offers an evolutionary advantage. For example, there is a moth - I''ve forgotten the particular species - that can change the colour of its body from light to dark when it''s environment has been burnt out. Clearly this adaptability offers an advantage, since birds find it very hard to find a dark bodied moth on the surface of a burnt tree. Those lighter bodied moths that couldn''t change colour would stand out and be eaten very quickly.


quote:
Original post by MikeD
The detachment from hard wiring in the brain might allow our bodies and brains to evolve more swiftly...



That''s an intersting thought. Although I''m not sure it is a detachment of hard-wiring, as opposed to a supression of hard-wired behaviours by cognitive decisions (supposed free will )... if that''s even possible of course!

Cheers,

Timkin

#53 Luctus   Members   -  Reputation: 580

Like
Likes
Like

Posted 14 July 2003 - 03:29 AM

quote:
Original post by Pipo DeClown
No one can create Intelligence, that''s why we have Artificial Intelligence.


I take it you''re not married?



-Luctus

Statisticly seen, most things happens to other people.
[Mail]


#54 Anonymous Poster_Anonymous Poster_*   Guests   -  Reputation:

Likes

Posted 15 July 2003 - 02:15 PM

I thought true Intelligence of a non-biological machine was defined as a machine that could pass the Turing Test. The Turing Test can be passed if a human on the opposite side of a "wall" as you can''t tell the difference between you and another human. The judgement would be based on intellect and not physical graphics. You is the bot.

#55 Timkin   Members   -  Reputation: 864

Like
Likes
Like

Posted 15 July 2003 - 03:58 PM

Stevan Harnad has proposed a graded series of Turing Tests for the assessment of intelligence, called T1 to T5. The original TT falls in at T3. The ultimate test, whereby the agent is completely indistinguishable from a human in all respects, is T5. It is postulated that the only way to pass T5 is to essentially be human - grow, learn and behave as a human does during their life - even though you might be made from artificial cells! The apparent requirements for passing T3 are only that the agent can absorb and understand the types of experiences that humans have and that they''re sensory systems are human-like; so that their underlying understanding of the things they are communicating about is human-like. This again gets back to what Mike has been talking about earlier in this thread.

Timkin

#56 Stonicus   Members   -  Reputation: 157

Like
Likes
Like

Posted 17 July 2003 - 07:25 AM

Putting humans at the top of the Intelligence ladder is rather arrogant... Better to make a scale T1 - T100 with humans maybe around T10 or something...


#57 RPGeezus   Members   -  Reputation: 216

Like
Likes
Like

Posted 17 July 2003 - 07:50 AM

quote:

Putting humans at the top of the Intelligence ladder is rather arrogant... Better to make a scale T1 - T100 with humans maybe around T10 or something...



I think we have to put humans at the top of the intelligence test-- not because we''re neccesarilly the most intelligent species in the universe, but because it would be impossible for us to comprehend the reasoning behind the actions of something vastly superior to humanity.

Will

#58 Timkin   Members   -  Reputation: 864

Like
Likes
Like

Posted 17 July 2003 - 04:40 PM

quote:
Original post by Stonicus
Putting humans at the top of the Intelligence ladder is rather arrogant...



No, it''s rather anthropomorphic...

The point of the Turing test is NOT to test for objective level of intelligence, but rather to test for the indiscernibility of an artificial agent and a human. Hence, a human SHOULD be the ultimate comparison test.

Timkin

#59 Boar_Volk9   Members   -  Reputation: 122

Like
Likes
Like

Posted 17 July 2003 - 04:56 PM

Wow, only just read this thread, its amazing the amount of differrent perspectives such a niche of humanity can have. i gotta say, like the rest, i too agree with a lot, and dissagree with a lot.
First off, as others have said, it is important to define your terms, and I think intelegence is just the abilty to do things. Plain and simple. A calculator is just as intelligent as a bot from counter-strike (yes I understant people hold contrasting views to this). Thus we can and have created artificial intelegence.
Many people are comparing intelegence to what the human race is, and can do. So my definition supports this.
I believe the original poster''s definition of intelligence may have been similar to my definition of ingenious. Which is the ability to invent, to hack, to look from different perspectives, to be irrational(yes i mean this one), to be able to change thier mind, and to be able to do choose not to go for the goal(weird? maybe too complex?(feedback please)). This definition of Ingenious however does not support the human race entirely, there are some among us that are ingenious, but it is almost infinitesimal.
I am second year uni student, studying IT and Maths, and wish to study AI thoroughly(though have study little yet), i believe that the true test comes when trying to develope something ingenious. Thus i repeat the original post in my own words(taking it a step further)...
Is it possible to create something ingenious?


#60 haro   Members   -  Reputation: 502

Like
Likes
Like

Posted 17 July 2003 - 05:19 PM

quote:
Original post by Zephyrox
The discussion here is about whats intelligence and whats not, and i dont belive anyone has an answer for that. Is a chess computer smart or intelligent? probobly not, it only has simple mathmatical rules to follow.
You cant tell if you playing against a computer or a human if your playing chess and cant see the opponent. So wouldn´t that be "real" AI?


Actually that''s a pretty bad analogy. Its near trivial to determine if you''re playing against a computer or a human, without seeing your opponent. You can also usually determine the approximate experience and strength of your opponent based solely on their moves in chess.

The strongest computers play really nothing at all like humans, and that isn''t to say they have a deeper understanding. Much the opposite computers tend to play extremely superficially by grand master standards, but they are extremely accurate at short term calculation.




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