Advertisement

How close are we?

Started by December 05, 2005 03:53 PM
33 comments, last by Madster 19 years, 2 months ago
Quote:
Original post by Fixxer
There will never be "REAL" AI.

Well, that's okay then. Debate's over, guys.
Quotes and answers shuffled to fit my argument ^^
Quote:
Original post by RPGeezus
It's not imagined. I can't prove this but you are aware of yourself, which you know, so it's not imagined.

I can't prove the world is flat, but you can see it for yourself, so it's not imagined. Lack of opposing evidence does not constitute a proof.

Quote:
Original post by Devilogic
Ok, I will accept your challenge. Here is a sketch of my proof of non-nondeterminism:
I will assume that I live in a world of complete freedom (and that the statement F "I am completely free" holds true by definition).

You define what you're trying to prove as an assumption. Thus, everything else you conclude will always be true. This isn't a proof either. And yes, formality is important, else anyone could claim things like the above.

Quote:
Original post by frob
I define it is the ability to make a choice between at least two options (aka Freedom model). That means you must have at least two options and (somehow) be able to take one or the other at your own free will.

Okay, this assumes somewhat the definition of 'choice' which hasn't been quite defined either.

What is choice?
I would define it as picking one option over a set of finite options based on a criteria. More on this after the next quote:
edit: dictionary says:
-The act of choosing; selection.
-A number or variety from which to choose
-The best or most preferable part.
we'll be refering mostly to the act of choosing, but the rest applies as well.

Quote:
Original post by alvaro
I will explain my personal position. I honestly think that there is no such thing as free will in a strong sense. That doesn't mean that people don't make decissions. It just means that decission making falls within the laws of physiscs, like everything else.

The problem of conciousness is probably not a problem at all. I think the problem is with the way humans tend to think of the world.
....
"conciousness" is probably an artifact of human [poor] thinking, and not an element of reality.

Two VERY important points are raised here, and only one of them was adressed and then only dismissed.

The first point is that decission making falls within the laws of physics. That is, choice is bound to physycs, like everything else.
Following that we could say that if physics were really deterministic, then choice would be deterministic, and if they were non-deterministic, so would choice be.
Then, from the previous definition of choice, if we can prove that the picked option over the set of possible options is always the same over different attempts with the same conditions, then we have proven that the universe is deterministic. ¿Still with me? well that's bad, because this is incorrect. :)
choice is bound by physics, so it is ruled by a subset of physics which may be or may not be the complete set of rules. So proving that choice is deterministic would only prove that some rules of physics are deterministic.

So, fully knowing this, I'll go on about choice. The thing is that choice is usually considered as 'free will'. But what makes one choose one option over another? could be your current mood. Could be the weather. Could be a passing memory. What's the point? all those things are part of the physical state!
So, when we say "given the same conditions", those are conditions too! so, given that the option is chosen based on these seemingly random factors and other more obviously deterministic factors, then the chosen option always ends up being the same.
Note, I've only said that freedom is an illusion, not that the universe is non-deterministic. Which is cool, because we were talking AI to begin with.

The second point alvaro raises is about consciousness being an artifact on human thinking. I'll just rant about this in a non-proof fashion.
First, we should acknowledge that we don't understand the whole universe. This is easy, since we have science to study the parts of it that we can perceive but not understand.
Then, we should acknowledge that we just cannot understand the whole universe, ever. This won't go down well with most people, but it's my personal perception.
Anyway, our model of the world isn't complete, yet we consider it intelligent.
So any top-down artificial intelligences we could come up with would be bound by our own intelligence, which is a slanted, limited view of the real world. Since we don't even understand how our own intelligence works, it follows that anything we could create trying to imitate it will be imperfect.

So, lacking any decent definition of intelligence I'd use this one:

The process behind choices and automatical improvement of said choices, so that a goal can be accomplished within the perceived world.

Actually I kinda like it. I'll jot it down.
Working on a fully self-funded project
Advertisement
Quote:
Original post by Madster
Quote:
Original post by Devilogic
Ok, I will accept your challenge. Here is a sketch of my proof of non-nondeterminism:
I will assume that I live in a world of complete freedom (and that the statement F "I am completely free" holds true by definition).

You define what you're trying to prove as an assumption. Thus, everything else you conclude will always be true. This isn't a proof either. And yes, formality is important, else anyone could claim things like the above.


My post was based around the idea, that it could be shown that any philosphical theory containing such an extreme degree of freedom would be incosistent with itself. I tried to illustrate this by considering what that degree of freedom would actually entail/what it would have for it's consequence. I believe the method I used is called proof by contradiction.

I'm beginning to see, though, that this isn't really the place to discuss (in joking or otherwise) such things, because people tend to take these things much too seriously. My post was never meant to be a solid proof, it was just a thought experiment... Maybe I sould just delete it and end this madness.

Quote:
Original post by Madster
Then, we should acknowledge that we just cannot understand the whole universe, ever.


One last comment: depends very heavily on what you mean by "understand the whole universe".
Conciousness.

Somewhere along the line of this thread I stopped reading accurately each and every post, so please don't blame me for barking in the air a bit. It's a friday night after all, and I've started drinking a bit already. ;-)

I've seen lots of 'technically complicated' language regarding the concept of conciousness in this thread. I've did some (humble) thinking about this too, and like to share my thoughts.
I am a believer of evolution theories and such, and therefore tend to think that all abstract concepts with which we tend to tackle the concept of conciousness are really besides the issue. I do not think that conciousness just 'came with the wind' of 'human intelligence'. Instead, it must be just a natural part of lots of animals. Most beings do not posses the intellectual concepts that we take for granted to express our conciousness, let alone that they are able to communicate about it's marvels to us. Things must be actually (relatively) rather simple instead. In terms of a system's approach (as a paradigm) conciousness might have to be not much more than a simple feedback-loop; a part of what goes out comes in again. It's like sensing the movement of your limbs, or being aware of your emotions; which are the strongest motivator for all sorts of behaviour (even in much less intelligent forms of life). Our 'real' conciousness is experienced at a more abstract level I imagine.

How this feedback of conciousness is realized in the fabric of our brain is a much more challanging question in my opinion. When the computer-era started, scientist were convinced they would be able to model the human brain in a computer soon. They were wrong obviously. The promise of the science of artificial intelligence was never really kept indeed...

Gotta go, my guest has arrived. :-)

Quote:
Original post by Devilogic
My post was based around the idea, that it could be shown that any philosphical theory containing such an extreme degree of freedom would be incosistent with itself.
ah contradiction. Yes, I didn't catch that. I could argue about some other parts then, but since you say it's not serious, I won't :)

Quote:

I'm beginning to see, though, that this isn't really the place to discuss (in joking or otherwise) such things, because people tend to take these things much too seriously. My post was never meant to be a solid proof, it was just a thought experiment...
Not really, we're supposed to talk about game development. Nothing wrong with a lil food for thought, though. Also, I'm used to follow trough to everything I say for a conclusion, and tend to assume everyone else wants to do that to.

Quote:
One last comment: depends very heavily on what you mean by "understand the whole universe".

Hummm it was late at night... what was the idea?
oh yes. Understand everything in the world/universe (everything) as it really is. In my view, that's impossible.
Working on a fully self-funded project

This topic is closed to new replies.

Advertisement