Darwinbots Forum

General => Off Topic => Topic started by: shvarz on May 13, 2005, 01:26:02 PM

Title: Definition of intelligence
Post by: shvarz on May 13, 2005, 01:26:02 PM
I wrote a post for LJ, which led to some interesting discussions, so I figured I may translate it to English and post here to see what you guys think about that.

We, humans, certainly believe that we are intelligent.  It is a phylosophical question whether we are, but we undoubtedly posess some of the attributes of intellligence: language, use of tools, cultural and social institutes and so on...  I can argue that these attributes and institutes are result of evolution and that micro-societies (tribes?) that did not posess them were more likely to die out.

Now if we look at DBs, we can see that (at least in theory) similar institutes can be developed and evolved.  The sims are simple right now, but I have already seen disappearance of cannibalism in my sim.  Given enough complexity we can imagine situations in which almost any our social institute has an advantage.

Now let's get back to intelligence.  We wouldn't call bots intelligent simply because they are not cannibalistic.  But what if they have complex language capable of describing abstract concepts?  What if they build houses?  Have kindergardens?  Write books?  Dance? Pray?  Make computers and run their own artificial life sims?  Would you at some point call these bots intelligent?  Don't answer yet, it gets a bit more complex.

DBs (and any other art life sime) is a deterministic system.  Give it the same seed for randomizer and the sim will repeat itself perfectly (there are more complex ways to introduce randomness, but let's not go there yet).  Now imagine that you were talking to a bot about the meaning of life and had a perfectly reasonble conversation.    However, if you start the sim over and ask the bot the same questions at exactly same point in sim, then you will get exactly same answers.  It becomes obvious that you were not talking to an intelligent being, you were talking to a very complex program.  Give it inupt "A" and it will always return "B".  That is not intelligence.

But why?  What is different?  Well, we intuitively feel that we, humans, on any given input "A" can give a vast number of outputs.  The person asking the question has no way to know in advance, which answer we give until we do.  So is the answer - unpredictability?  But unpredictability by whom?  Bots can't predict each other's answers, they don't have enough information on how DBs works.  But they are predictable to us.  So, we need to make them unpredictable to us, humans.  How?  Well, we can hook up a Geiger counter to computer and seed a truly random number into the sim at each and every cycle.

Ingenious?  No, it is not.  Think about it - if before you were talking to a very complicated program, now you are talking to a very complicated generator of random answers.  Right?  10 electorns hit geiger counter, the answer is "B", 20 electrons - "D".  

This is not intelligence.

In order to consider the oponent intelligent we must assume that it is capable of giving many different answers, but it does not choose the answer randomly.  This is described by the concept of "free will".  The problem, however, is that "free will" is a phylosophical concept and does not have a measurable manifistation in our world.  We can only state that we feel free will in ourselves, but we have no clue about others.

As a reuslt, our definition of intelligence relies not on some scientific observation, but on a belief in existence of free will in other people.  We believe that they have free will, they believe that we have free will.  Mutual agreement to stabilize our psyche.  The most basic and, probably, the most ancient sociological institute in human history.

So, where does it leave us?  Still in the same place - intelligence is undefinable by scientific methods and is a matter of opinion only.

Ironic, isn't it?
Title: Definition of intelligence
Post by: PurpleYouko on May 13, 2005, 01:51:31 PM
Oh crap!!

That means that if we do indeed live in a deterministic universe then none of us are actually inteligent at all.

Makes you wonder doesn't it?

If the universe could be started all over again, in exactly the same way with identical starting conditions, would everything turn out exactly the same? Down to every word said by every individual throughout all time.

Or would there be variations?

And how big would they be?

In the reproduced universe, would I just be wearing a different colored shirt today or would the whole world be populated with quasi-inteligent descendants of Trilobites that look somewhat like purple Twinkies with 18 legs and 6 eyes?
Would I then be writing this same post but speculating about some big hairy biped that might have become the dominant species if things were started over?

 :blink:

Now look what you have done Shvarz. You went and made me get all philosophical and mind boggly. Now I will never get any sleep tonight.
Title: Definition of intelligence
Post by: Numsgil on May 13, 2005, 02:52:14 PM
I think that on the quantum level true unpredictability exists.  It may be that we just don't understand the rules behind it, but then again, it may really be unpredictable.

Now, in our billions of years down the line world, I don't think quantum fluctuations would matter.  But in the first 2^-32 seconds, these could matter alot.

Also, in the interactions of the brain, quantum fluctuations could matter alot.  It may be that which makes us choose one response over another, it's hard to say.

You have to view the brain as a possibly very complex black box machine that takes inputs and gives outputs.  Unfortunately, inputs also change the machine, so it becomes impossible to understand the inner workings of the machine.
Title: Definition of intelligence
Post by: Botsareus on May 13, 2005, 03:06:04 PM
Shvartz, you are saying that we are intelligent just because we cant understand how the universe works that makes us intelligent. right?

What if we figure it all out, then we are not going to be intelligent anymore? or are we going to be more intelligent?

The Point is do not make up complicated definitions for something simple, if you like a person or being you call it intelligent , if you don’t like it , you don’t call it intelligent that’s all.
Who is right? Well that depends were you going to apply the fact that n is intelligent (if you had to choose from an array of n for example). If for you the correct filtering out of n worked. Then you would say that definition of intelligent is good.
Title: Definition of intelligence
Post by: shvarz on May 13, 2005, 03:08:22 PM
Two thigs Nums:

1.  The fact that something is unpredictable does not mean that it does not follow "cause-effect" rule.  At any given time the state of universe A is converted to state of universe B through some function F (rules of universe).  The fact that we don't know function F (even can't know the function) does not mean that the function does not exist.

2.  If quantum effects indeed rule our brain, then our decisions are either a) random or B) predictable (see point 1).  In either case it is not us who is making decisions, it is the universe as a whole.
Title: Definition of intelligence
Post by: MightyPenguin on May 13, 2005, 03:36:53 PM
I'm not sure about intelligence, but I came up with three criterion for sentience off of the top of my head.

1) The being in question must be capable of abstract thought.
2) The being in question must be able to understand that it exists.
3) The being in question must be able to recognise itself as an individual entity.

Those might be a bit ambiguous...
Title: Definition of intelligence
Post by: PurpleYouko on May 13, 2005, 03:44:32 PM
Quote
In either case it is not us who is making decisions, it is the universe as a whole.
That is a bit of a Zen way to look at it.  B)

So what you are really saying is that we are all just momentary ripples in river of existence. During  the lifetime of each ripple it is breifly subjected to the illusion of self awareness but it soon merges with the collective conciousness of the river again.
In this there is no life or death, just an unceasing change of the forms of consciousness.
Title: Definition of intelligence
Post by: shvarz on May 13, 2005, 04:07:39 PM
Quote
So what you are really saying is that we are all just momentary ripples in river of existence. During the lifetime of each ripple it is breifly subjected to the illusion of self awareness but it soon merges with the collective conciousness of the river again.

I am saying that it is pretty much impossible to tell whether it is true or not.  At least at this point.

MP: regarding your three points:
How are you going to judge your numbers 2 and 3?  Ask the creature?  What if it is programmed to answer whatever you consider to be "correct" answers.
Title: Definition of intelligence
Post by: MightyPenguin on May 13, 2005, 04:32:15 PM
Quote
MP: regarding your three points:
How are you going to judge your numbers 2 and 3?  Ask the creature?  What if it is programmed to answer whatever you consider to be "correct" answers.
It would have to be proved beyond resonable doubt. I'm not even going to pretend to know how to do that.

I'm simply saying; you'd have to convince me of those three before I would accept something as sentient. How you choose to prove it is up to you.

I'm going to back out now, before someone proves I'm just a figment of my own imagination.

Damn philosophy.
Title: Definition of intelligence
Post by: shvarz on May 13, 2005, 04:43:21 PM
Well, my point is that the formal proof of intelligence is not possible.  Therefore, you will have to believe that something is intelligent.
Title: Definition of intelligence
Post by: MightyPenguin on May 13, 2005, 04:48:11 PM
Quote
Well, my point is that the formal proof of intelligence is not possible.  Therefore, you will have to believe that something is intelligent.
I was just answering the topic title. To be honest with you, I can't be bothered reading posts longer than four lines on this place anymore.
Title: Definition of intelligence
Post by: Anonomous Guest Person on May 13, 2005, 05:57:41 PM
How 'bout this for a definition of intelligence?

"Whatever can, even if it's incapable of showing it, prove this definition illogical is intelligent."

Practical, and not logical at the same time!
Title: Definition of intelligence
Post by: Numsgil on May 14, 2005, 12:47:51 AM
Well, as Schvarz mentioned, we don't even know that each other is intelligent. We just assume so since they seem to be able to communicate in a manner that would lead us to that belief.

So the Turing test may be the only true test of intellegence.  If it seems intelligent, it is intelligent.  Intelligence may be osmething that can only be defined recursively.

Intelligence is anything that appears intelligent to an intelligent creature.  I know I'm intelligent.  All other (okay, most other) humans seem intelligent to me, so they are now defined as such.  If another human encounters something I haven't seen yet and claim it is intelligent, then it too is intelligent.

An interesting idea.
Title: Definition of intelligence
Post by: Botsareus on May 14, 2005, 11:21:45 AM
Thats what I was getting at.

 :D Bau
Title: Definition of intelligence
Post by: Numsgil on May 15, 2005, 01:17:55 AM
Okay I had this idea:

Free will would seem to be either:

1.  An illusion of a deterministic universe -or-
2.  Based in true random quantum events.

'Free will' is basically being able to do something that's unexpected.  If I can predict what you'll do from any set of inputs then you're just a machine.  So this unpredictability may very well come from a true randomness generator - that of quantum probabilities and energy levels.

Or it may be that your head is indeed a finite state machine, but one that changes it's internal probabilities after it gives an output.  That is, psuedorandom, like the way computers generate random numbers.

There may be a third possibility, but I think this is what free will and sentience basically is - a way for the universe to amplify the micro-probabilities that don't have very large macro effects.  So sentience could be seen as the univere's extension of unpredictability.
Title: Definition of intelligence
Post by: shvarz on May 16, 2005, 01:16:02 AM
We experience the free will all our lives and your explanations don't explain this fact.  Or at least they don't explain it in any reasonably simple sort of way.  The irony here is that in this case the scientific approach leads us to a non-scientific conclusion, I.e. that there is such a thing as free will and it has an impact on the way this universe functions.  "Look, I will to move my hand and now all these atoms are moving in a direction that I wanted them to move!"

If we go down to quantum physics then maybe we will find that at some point the will of an observer can change the outcome of a quantum experiment.
Title: Definition of intelligence
Post by: Endy on May 17, 2005, 06:16:06 PM
I think I read something about that it's better to behave as if we have free will than to not, because even if we don't it has already been determined that we would.

It is possible to reasonably predict how another person will act/react, if enough is known. I don't see how that is really all that different from knowing how a machine will behave.

Endy B)
Title: Definition of intelligence
Post by: shvarz on May 27, 2005, 02:02:00 PM
If anyone is interested in this topic, then here is a linky to a set of very interesting lectures about relationship between modern physics and free will: http://www.consciousness.arizona.edu/quant...gi?msgindex.mif (http://www.consciousness.arizona.edu/quantum/Archives/Lectures/mifdex.cgi?msgindex.mif)

The lectures are of different level, some are more techical, others - less, but overall a very intersing discussion.