Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
7 online now:
Newest Member: popoi
Post Volume: Total: 915,815 Year: 3,072/9,624 Month: 917/1,588 Week: 100/223 Day: 11/17 Hour: 0/7


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Why doesn't AI Falsify ID?
crashfrog
Member (Idle past 1466 days)
Posts: 19762
From: Silver Spring, MD
Joined: 03-20-2003


Message 61 of 71 (374167)
01-03-2007 7:36 PM
Reply to: Message 60 by Percy
01-03-2007 7:30 PM


Right, that's what I was going by.
I mean, I guess it's up to him to tell us, but when I substitute actual words for where he uses X and Y, I get:
I assert that even if "CSI is detectable" was true, it would still not imply "detecting CSI allows us to infer non-physical causation".
It's a little clumsy, but from that I understand him to be asserting that his argument is that CSI doesn't imply "non-physical causation" regardless of its detectability.
But I guess you're right, now that I think about it. If CSI doesn't have any meaning, we can't say what it does or doesn't imply. It has to be defined, first. Doesn't it?

This message is a reply to:
 Message 60 by Percy, posted 01-03-2007 7:30 PM Percy has not replied

Replies to this message:
 Message 64 by aiguy, posted 01-03-2007 10:50 PM crashfrog has replied

  
aiguy
Inactive Member


Message 62 of 71 (374254)
01-03-2007 10:39 PM
Reply to: Message 57 by Hyroglyphx
01-03-2007 6:38 PM


Re: Computer models
Hi Nemesis,
Nemesis writes:
You're probably right that we'd receive different answers, but at the same time, if we were given examples, I think we'd all generally agree.
Yes, intelligence is famously like pornography (according to Ed Meese): Can't define it, but we know it when we see it.
Nemesis writes:
Meaning, suppose we saw the scores of a person after they've taken an IQ test. Because we know that the test was difficult coupled with the yielding of a high score, most people would generally come to the conclusion that he/she is highly intelligent.
The problem is we can't give IQ tests to chimps or ravens or cells or the Intelligent Designer to figure out if they're intelligent or not.
Nemesis writes:
But understand what you are really getting at. You are essentially saying that we can't really quantify intelligence and define it in a meaningful way kind of like we couldn't describe love scientifically in a meaningful way.
Just so. Doesn't mean these words have no meaning, just not scientific meaning.
Nemesis writes:
AIGUY: That is, something with "will" can break the known laws of physics and make something happen that has no antecedent physical cause. Well, there is no evidence that anything of the sort ever happens.
NEMESIS: I have no idea what you're talking about. Can you please expound?
Intuitively, people feel that what they decide in their mind causes their bodies to do what they do - "mind over matter" and all that. However, if the world is nothing but deterministic cause-and-effect, then it seems we don't really have any power over what we decide - it has already been decided for us by the laws of physics. One school of thought called "libertarian" or "contra-causal" free will holds that our minds can actually interfere with physical processes from outside, as it were. Others look to quantum indeterminacy to support the idea that mind can act in the physical world without breaking the laws of physics (this has not been very successful, in many people's opinion). Other people (like me) are "compatibilists" - those that believe our actions probably are determined by physics, but for any important meaning of "will", it doesn't matter.
Nemesis writes:
AIGUY: Computer systems do things that their programmers have never anticipated.
NEMESIS: Such as what? Is this all computers or supercomputers?
Supercomputers are just computers - same deal, just bigger and faster. I'm talking about systems that 1) use randomness and selection, making their actions unpredictable in principle, and 2) those that are so complex and heuristic that the programmer can't possibly ever trace the results of execution without actually running the program.
Nemesis writes:
How can something synthetic go through any kind of natural process? Also, in order for necessity to exist, there first has to be potential consequences. What consequence exists for computers if they didn't evolve?
My point is that computers act in accordance with blind, unguided physical laws. Yet they still act design complex machines. Thus, blind unguided physical laws in nature (such as evolutionary processes) can design complex machines (living things) too.
Nemesis writes:
I do believe that based on any number of things. Most notably is that its counter intuitive to suppose that anything can come into existence without causation. However, that's just my opinion on the theory. Its one of those paradoxes that neither stands strong or falls flat on its face. There will likely be a perpetual stalemate on the subject.
Just so. So we leave ultimate cause out of science, and concede that science doesn't explain everyting.
Nemesis writes:
I don't think anything other humans possess human-like emotions.
Well, we disagree there (I absolutely believe my dogs are happy, sad, frustrated, eager, embarassed, etc). But again, it is very hard to see how we could ever test to see if I'm right.
Nemesis writes:
What I am saying is, the more intelligent an animal is, the greater capacity they have for emotion. Intellect is very far reaching. I doubt that it only incorporates book smarts or an ability to engineer. Case in point: Would most people be likely to esteem Beethoven as intelligent? They probably would, but why?
I absolutely agree. Antonio D'Amasio and others have changed cognitive science in the past twenty years or so by explaining now emotions are intrinsic to human decision-making. And I would say that a very small proportion of our knowledge comes from "book smarts". But my point was that if we observe the results of intelligence - as ID would say that biological systems are - we have no justification for inferring emotions along with it, or any other aspect of human mentality.
Nemesis writes:
Then we're left with only mechanical function to equate with intelligence, which makes calculators intelligent.
Calculators do what they do. If you want to call them intelligent, or not, it doesn't make any difference.
Nemesis writes:
This machine learning and self-programming, what does it entail? And what kind of software is needed for a computer to think for itself and not just choose the command it was given?
There are various techniques for ML. Basically the software is set up to analyze its experiences and look for patterns and correlations, and then adjust its behavior accordingly.
Nemesis writes:
Your link didn't direct me to anything specific. But as for computers and common sense, isn't the mere fact that they can only perform the functions assigned to them prove that they aren't actually intelligent?
We don't seem to making progress on this point. Once again:
1) They can perform functions that have not been assigned to them
2) It doesn't make any difference whether or not we call them intelligent
Nemesis writes:
AIGUY: Do you think we ought to give this theory a scientific chance in order to see if it's true? (Don't ask me what gunderplitzen is - I'm not really sure yet).
NEMESIS: If you developed a thesis on it, yes, of course. But you obviously use jocularity to make up a word that has no definition. It would be extremely unfair to characterize ID with that of gunderplitzen. If ID had no backbone, we wouldn't see many people in abject terror of it. Afterall, people don't fear straw men.
First, yes it was a joke. My point was that not everything can be scientifically evaluated. In order for something to be scientifically evaluated, the "thesis" must be coherent, and make specific, meaningful statements about what we can observe in the world. I do not believe ID qualifies, because the word "intelligence" has not been given a suitable operational defintion.
Nemesis writes:
How it became intelligent is the only way to answer any questions. Again, science would die without those burning questions. Understanding the intelligence of the programmer is going to give you a greater understanding of the intelligence capacity of the computer program itself. One could apply that rationale to the Designer(s) of the universe/multiverse. Of course, answering that is exceedingly difficult-- granted.
You're missing my point. Of course we'd like to learn all about causes as far back as we can. But that isn't the point of my argument. The point is that AI computers prove that pure, blind, chance and necessity is capable of building novel complex machines, and that argument is not countered by pointing out that human beings built computers.
Nemesis writes:
Who says that it has to be meaningless, is my question? Is language meaningless?
What I'm saying is that explaining biological complexity in terms of "intelligence" is meaningless (or vacuous):
Q: What causes biological complexity? A: Intelligence.
Q: How do you know? A: That is what we call things that can generate complexity.
Nemesis writes:
A game on my computer can have rules. Say the computer based on my movements deduces that I fall off of a cyber cliff if I stray to far from the edge. The programmers designed it the away though, not the computer.
Computers can make up their rules, using induction.
Nemesis writes:
AIGUY: deduce facts from other facts, diagnose strange medical disorders
NEMESIS: Based on the previous input supplied by doctors and programmers.
And doctors learn from other doctors too (their teachers). So?
Nemesis writes:
Could a computer be intelligent without any input or influence of an outside source?
I'm going to try this just one more time:
1) ID claims that natural processes can't design complex machines.
2) Computers design complex machines
3) Computers act according to natural processes
4) Therefore (1) is false
You will now complain that computers can only design complex machines because they are designed to do so by humans.
I will rebut by pointing out it doesn't matter - humans can only design complex machines because they were designed to do so. And nature can only design complex machines (living things) because it was designed to do so. Now - before you get all excited and think that I've just adopted ID - I HAVE NOT. We have absolutely no idea how nature was "designed" such that natural processes could cause life to form. We have no scientific reason to think it had anything to do with a mind.
Nemesis writes:
It would seem to appear to be the case because have a chicken and an egg problem without it.
We just don't know.
Nemesis writes:
Necessity indicates intent. If you speak of nature in terms of itself gaining insight on self-preservation, then you are ultimately speaking about nature as if it is sentient.
No, I do not believe nature is sentient. But what I'd really like to hone in on is the point that ID's explanation of "intelligence" outside of nature for biology fails as a scientific theory. One simply cannot scientifically infer that anything outside of natural processes was involved.
Nemesis writes:
AIGUY: My point is that ID offers "intelligence" as the explanation for biology, but "intelligence" has no scientifically useful definition. That isn't my fault.
NEMESIS: Isn't intelligence the explanation for why your computer is intelligent?
Of course not!!!
Q: How does your computer play chess? A: Because it is intelligent.
Q: Why do you say your computer is intelligent? A: Because it can play chess.
In general, and in the case if ID, calling something "intelligent" does not tells us anything we didn't already know.
Nemesis writes:
In the same way people can have a theory on chance as the explanation.
I hope you are not saying evolutionary theory invokes "chance" as an explanation for biological complexity - it doesn't.
Nemesis writes:
Really what this is, is the typical mischaracterization of what ID is. For some bizarre reason, opponents of the theory think that if a Designer is invoked that somehow science will cease. What an absurd notion. Does it cease if we think that random chance is the answer? Obviously not. ID is nothing more than an inference that uses observation to corroborate the claim. I mean, really, what threat exists from it?
Nobody thinks chance explains biology. Also, we cannot have a scientific theory that explains things by invoking a Designer, because nobody knows what a Designer is. We cannot fix this by saying it is "intelligent", because as we've seen there is no scientific content to this assertion. And the threat is that people will try to use the authority of science to further their own particular opinions about the Designer that have no scientific support.
Nemesis writes:
You missed the rest of the post. If that truly is the case, then you would be incapable of discerning chance from design. That means you couldn't argue that design in nature is an absurd concept, because you would inevitably have to say that chance within nature is just as absurd.
I'm not sure I follow everything you've said, but I agree "chance" is a tricky concept. (There are really two kinds recognized by science - randomness caused by unfathomable interactions of contingent events (like the pattern of raindrops on the driveway) and quantum randomness, which is inherently undetermined). But evolutionary doesn't explain anything by chance except for individual mutations.
Nemesis writes:
No argument there. That's like supplying the answer to this question:
Why are there complex organisms in existence. Answer 1. Goddidit! Answer 2. Evolution! Neither explain squat without describing the how's and why of it. An answer of such brevity is inept to answer anything.
Well, no, it's not the brevity at issue here. If we say "random mutation and natural selection" did it, we really are making a claim with clear meaning, whether or not you think RM&NS (or any aspect of evolutionary theory) really accounts for living things or not. On the other hand, saying "intelligent causation" explains it is really not saying anything at all - it is neither right nor wrong, because it just doesn't say anything (in science we say it is "not even wrong").
Nemesis writes:
Its a pleasure having you here. I hope you find it intellectually stimulating and spiritually fulfilling here at EvC.
Thank you so much. I've enjoyed it so far - the only cross words I've encountered have come from somebody who actually agrees with me on the main issues of the ID/evolution debate! I enjoy your sincere and polite conversation.

Science is not simply reason - it is much less than that. It is reason constrained by empiricism.

This message is a reply to:
 Message 57 by Hyroglyphx, posted 01-03-2007 6:38 PM Hyroglyphx has not replied

  
aiguy
Inactive Member


Message 63 of 71 (374259)
01-03-2007 10:45 PM
Reply to: Message 58 by Percy
01-03-2007 7:11 PM


Hi Percy,
Percy writes:
I agree with your basic approach, but I just don't see how a successful argument can be built upon postulating the detectability of something with no formal definition. I think you'll ultimately bog down in digressions over what CSI really is, at which point you'll have to agree upon a definition, which brings you back to what I thought was the more significant issue, CSI's lack of any formal definition. But hey, give it a try!
Once I grant, arguendo, that complex form and function is detectable, we just remove that from the discussion. Actually, where people bog down is that they can't understand why the fact that programmers program computers doesn't invalidate the argument.

Science is not simply reason - it is much less than that. It is reason constrained by empiricism.

This message is a reply to:
 Message 58 by Percy, posted 01-03-2007 7:11 PM Percy has not replied

  
aiguy
Inactive Member


Message 64 of 71 (374262)
01-03-2007 10:50 PM
Reply to: Message 61 by crashfrog
01-03-2007 7:36 PM


Hi Crashfrog,
Crashfrog writes:
I mean, I guess it's up to him to tell us, but when I substitute actual words for where he uses X and Y, I get:
I assert that even if "CSI is detectable" was true, it would still not imply "detecting CSI allows us to infer non-physical causation".
Yes, just so.
Crashfrog writes:
But I guess you're right, now that I think about it. If CSI doesn't have any meaning, we can't say what it does or doesn't imply. It has to be defined, first. Doesn't it?
The vast majority of IDers have no idea how Dembski has defined CSI (and he's changed it several times!). All people really think about is Paley's watch, and how it implies a watchmaker. My point is that we have observable proof that natural processes can build a "watch" - a complex functional machine.
Edited by aiguy, : No reason given.

This message is a reply to:
 Message 61 by crashfrog, posted 01-03-2007 7:36 PM crashfrog has replied

Replies to this message:
 Message 65 by crashfrog, posted 01-03-2007 11:09 PM aiguy has not replied

  
crashfrog
Member (Idle past 1466 days)
Posts: 19762
From: Silver Spring, MD
Joined: 03-20-2003


Message 65 of 71 (374268)
01-03-2007 11:09 PM
Reply to: Message 64 by aiguy
01-03-2007 10:50 PM


My point is that we have observable proof that natural processes can build a "watch" - a complex functional machine.
Point taken. I agree, obviously.

This message is a reply to:
 Message 64 by aiguy, posted 01-03-2007 10:50 PM aiguy has not replied

  
herrmann
Junior Member (Idle past 6141 days)
Posts: 11
Joined: 05-22-2007


Message 66 of 71 (403539)
06-03-2007 9:01 PM
Reply to: Message 1 by aiguy
12-30-2006 5:08 AM


you do not necessarily have to say that machanical processes are intelligent because they do not have a soul.
According to dictionary.com, intelligence is defined as " 1. capacity for learning, reasoning, understanding, and similar forms of mental activity; aptitude in grasping truths, relationships, facts, meanings, etc. ".
Understanding requires conscious thinking, and mechanical processes cannot reason outside of specific information imputted into them through programming. Also, these programs cannot think, but only have the illusion of thought.
An analogy of this would be an array of lightbulbs. By flicking on certain lightbulbs, you may be able to spell your name. You could give a computer instructions as to when to flick on certain lightbulbs, but it would never have a stream of conscious thought.

This message is a reply to:
 Message 1 by aiguy, posted 12-30-2006 5:08 AM aiguy has not replied

Replies to this message:
 Message 67 by crashfrog, posted 06-03-2007 9:15 PM herrmann has replied

  
crashfrog
Member (Idle past 1466 days)
Posts: 19762
From: Silver Spring, MD
Joined: 03-20-2003


Message 67 of 71 (403541)
06-03-2007 9:15 PM
Reply to: Message 66 by herrmann
06-03-2007 9:01 PM


Understanding requires conscious thinking, and mechanical processes cannot reason outside of specific information imputted into them through programming.
I would offer the human brain as an example of a mechanical process that is able to reason outside of information inputted by programming.
You could give a computer instructions as to when to flick on certain lightbulbs, but it would never have a stream of conscious thought.
Says you. Many contend that, indeed, a computer could have a stream of conscious thought. And indeed, if a computer could fool everyone into believing that it did, how would that be different? How can you prove that you're not just fooling everyone with an outward simulation of conscious thought?

This message is a reply to:
 Message 66 by herrmann, posted 06-03-2007 9:01 PM herrmann has replied

Replies to this message:
 Message 68 by herrmann, posted 06-04-2007 3:49 PM crashfrog has replied

  
herrmann
Junior Member (Idle past 6141 days)
Posts: 11
Joined: 05-22-2007


Message 68 of 71 (403631)
06-04-2007 3:49 PM
Reply to: Message 67 by crashfrog
06-03-2007 9:15 PM


quote:
Says you. Many contend that, indeed, a computer could have a stream of conscious thought. And indeed, if a computer could fool everyone into believing that it did, how would that be different? How can you prove that you're not just fooling everyone with an outward simulation of conscious thought?
you can prove that you are not using an outward representation of conscious thought by consciously thinking. If a machine does have conscious thought, it is unable to manifest it into action, and thus it is meaningles as it has no pupose.
By my understanding as long as the computer does not have the ability to truely understand this programming, it does not have intelligence.

This message is a reply to:
 Message 67 by crashfrog, posted 06-03-2007 9:15 PM crashfrog has replied

Replies to this message:
 Message 70 by crashfrog, posted 06-06-2007 7:55 PM herrmann has not replied

  
molbiogirl
Member (Idle past 2641 days)
Posts: 1909
From: MO
Joined: 06-06-2007


Message 69 of 71 (404137)
06-06-2007 7:33 PM


Another perspective on AI
Jeff Hawkins, in a talk he gave at TED in 2003, offers "a recipe for brain theory" and how it relates to AI.
I'll just let TED speak for Jeff for a moment:
To date, there hasn't been an overarching theory of how the human brain really works, Jeff Hawkins argues in this compelling talk. That's because we still haven't defined intelligence accurately. But one thing's for sure, he says: The brain isn't like a powerful computer processor. It's more like a memory system that records everything we experience and helps us predict, intelligently, what will happen next. Bringing this new brain science to computer devices will enable powerful new applications -- and it will happen sooner than you think.
Jeff takes something Francis Crick said in a September 1979 Scientific American article (to paraphrase, "We don't know squat about brains. We lack a framework to interpret the scads of data we have.") and runs with it.
He defines intelligence and then talks about his brain theory:
Jeff Hawkins: How brain science will change computing | TED Talk
He's a compsci guy and I don't want to mangle his ideas. It might be best to take a gander for yourselves (that link is a 20 minute video).
Caveat: It very well may be that since 2003 somebody picked up the gauntlet Jeff threw down. I dunno. I'd have to go take a look around to find out.

  
crashfrog
Member (Idle past 1466 days)
Posts: 19762
From: Silver Spring, MD
Joined: 03-20-2003


Message 70 of 71 (404144)
06-06-2007 7:55 PM
Reply to: Message 68 by herrmann
06-04-2007 3:49 PM


you can prove that you are not using an outward representation of conscious thought by consciously thinking.
I can prove it to myself, and you can prove it to yourself - but neither one of us can prove it to the other.
I know I can think. I know I'm using conscious thought. How do I know that you are? How do you know that I am?
Do you understand the problem, here? Just saying "I'm using conscious thought" doesn't prove it, because I could program a computer to say that. You could be just a computer, as far as I know. As far as you know, I am.
If a machine does have conscious thought, it is unable to manifest it into action, and thus it is meaningles as it has no pupose.
The fact that no computers appear to have conscious thought yet is not proof that they never will.

This message is a reply to:
 Message 68 by herrmann, posted 06-04-2007 3:49 PM herrmann has not replied

  
waqasf 
Inactive Member


Message 71 of 71 (462918)
04-10-2008 12:30 PM


Spam deleted.
Edited by AdminModulous, : No reason given.

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024