I forgot to mention:
I agree with Ken that there are fundamentals missing from our best
efforts to make an artificial intelligence. The "smartest" programs
are extremely specialized. What's missing is the software to let
programs become "smart" about arbitrary subjects - in other words, to
learn. The problem is that learning requires experience, and there's
no reasonable way to give software any real experience.
Moravec says it's easy to teach a computer checkers, but hard to get
it to move. That underscores the situation. Checkers is easy because
there are very few, very specific rules involved. No experience is
required. But to make up for a lack of experience in mobility, we
must program incredibly complex control systems. We're trying to make
up for the lack of experience with rules. It is, I should note, quite
easy to make a six-legged robot that quite literally learns to walk in
minutes. The problem is scaling the flexibility needed to do
something "simple" like walking to harder things that we take for
granted because we had the luxury of our first 18 years of life or so
to learn to do them.
Having said that, I also have absolutely no doubt that we'll be able
to create a real, artificial intelligence in the next century -
assuming we don't blow ourselves up first.
Cheers.
Fil
On 19 June 2011 16:41, Filippo A. Salustri <[log in to unmask]> wrote:
> I think being able to design (at least in Simon's sense) is something
> we evolved to do. There's examples of octopi opening jars and
> uncorking bottles to get at food. (e.g.
> http://www.youtube.com/results?search_query=octopus+opening+jar&aq=0&oq=octopus+ope)
> - If not designing, then there are elements of design at work here
> (planning, problem solving, etc).
>
> I think the feature that most distinguishes humans is that we *know*
> we're designing. I don't think animals like octopi & dogs reflect as
> we do.
>
> Cheers.
> Fil
>
> On 19 June 2011 09:02, Ken Friedman <[log in to unmask]> wrote:
>> Dear Marcio,
>>
>> Thanks for your thoughtful response to my note on embodied judgment.
>>
>> While I can see that it may seem that I link our ability to project and
>> judge to the notion of human mortality, the relation is simpler than
>> this, but more subtle. Our ability to judge and to project is a function
>> of being embodied. Our bodies are mortal, and in this respect, empathy,
>> understanding, and many other aspects of what it is to be human are
>> predicated on mortality and empathy for the sufferings and pleasures
>> which all mortals experience. If we had immortal bodies, then the forms
>> of empathy, understanding, and what it might be to be human would be
>> predicated on immortality. It is interesting in a speculative sense to
>> wonder what forms of suffering and pleasure we would experience as
>> immortals.
>> [...]
--
\V/_
Filippo A. Salustri, Ph.D., P.Eng.
Mechanical and Industrial Engineering
Ryerson University
350 Victoria St, Toronto, ON
M5B 2K3, Canada
Tel: 416/979-5000 ext 7749
Fax: 416/979-5265
Email: [log in to unmask]
http://deseng.ryerson.ca/~fil/
|