Re: Why human colonization may be irrelevant
- From: ZnU <znu@xxxxxxxxxxxx>
- Date: Wed, 04 Jan 2006 01:46:46 -0500
In article <6dHuf.30444$X25.576902@xxxxxxxxxxxxxxxxxxxxx>,
Alain Fournier <alain245@xxxxxxxxxxxx> wrote:
> ZnU wrote:
>
> > Even intelligence, which initially seems like it might be easier to
> > identify, is a slippery thing. We've got computers that play chess
> > better than any human alive. Are they intelligent? How about systems
> > which can identify objects visually? Are those intelligent? Many people
> > will say something along the lines of "No, it's not real intelligence,
> > because the software just does X..." But one day, we'll be able to say
> > the same kinds of things about the human brain. "It doesn't really
> > require intelligence to write a symphony, because your brain just does
> > X..."
> >
> > Anyway, what I'm getting at here, is that machine intelligence and/or
> > machine consciousness probably won't be the result of a single
> > breakthrough. They could very well be the result of an evolutionary
> > process, with lots of gradual improvements.
> >
> > Of course, those gradual improvements will probably take a long time.
> > So, your point is certainly a good one, that we could colonize space now
> > with enough money, while we couldn't build a machine intelligence now,
> > no matter how much money we tossed at the problem.
>
> I think you are underestimating how well we understand the human brain.
> We don't fully understand it, but we do understand a lot about it. I do
> research in psychiatry for a living. Though I don't personally study
> the inner workings of the brain, I do attend to seminars on the
> subject every now and then. I saw a guy demonstrate a computer
> program that could identify if there was a face in a picture. Of
> course such programs have existed for a while. What was peculiar
> about his program was the way it was built. The programmer didn't
> try to write code to recognize faces, he wrote code to simulate
> the behavior of about a dozen different kind of brain cells. That
> was enough to recognize if there was a face in a picture. That
> computer program could probably be easily modified to perform
> other tasks by changing the set of patterns stored. If instead
> of storing eye, nose and mouth patterns you store patterns of
> linear and quadratic functions you might get it to solve some
> math problems.
Sure, basic pattern recognition like that can be done with very
simplistic neural networks. I played around with a simple version of
this trick years back, when I was teaching myself how to program.
Humans do more than just recognize patterns, though. There appear to be
all sorts of complex feedback loops that exist within the brain as well.
> Most (all?) human thinking involves recognizing patterns and checking
> if all the pieces are at the right place and then if the pieces are at
> the right place then a cell sends a signal saying we have a match.
> What happens with the signal saying we have a match, some times that
> signal is sent to a muscle to make it flex other times the signal
> is sent to another part of the brain to see if that match will
> form another pattern with other stuff in the brain. So if we can
> understand how it is done for finding faces we understand a lot
> about the human brain.
--
"Those who enter the country illegally violate the law."
-- George W. Bush in Tucson, Ariz., Nov. 28, 2005
.
- References:
- Why human colonization may be irrelevant
- From: alexandersheppard
- Re: Why human colonization may be irrelevant
- From: Rüdiger Klaehn
- Re: Why human colonization may be irrelevant
- From: ZnU
- Re: Why human colonization may be irrelevant
- From: Alain Fournier
- Why human colonization may be irrelevant
- Prev by Date: Re: Watch The Great Conspiracy on Google Video
- Next by Date: Re: Why human colonization may be irrelevant
- Previous by thread: Re: Why human colonization may be irrelevant
- Next by thread: Re: Why human colonization may be irrelevant
- Index(es):
Relevant Pages
|
