Re: Aaron Sloman's "The Irrelevance of Turing Machines to AI" article

From: Sergio Navega (snavega_at_intelliwise.com)
Date: 08/02/04 

Date: Mon, 2 Aug 2004 17:31:46 -0300

"Neil W Rickert" <rickert+nn@cs.niu.edu> escreveu na mensagem
news:ceir6u$rt4$1@usenet.cso.niu.edu...
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> "Sergio Navega" <snavega@intelliwise.com> writes:
>
> >"Neil W Rickert" <rickert+nn@cs.niu.edu> escreveu na mensagem
> >news:ceel1a$vvq$1@usenet.cso.niu.edu...
> >> -----BEGIN PGP SIGNED MESSAGE-----
> >> Hash: SHA1
>
> >> "Sergio Navega" <snavega@intelliwise.com> writes:
>
> >> >I still don't understand why this has to be so. I can think of
> >> >a computational specification where there's no a priori true or
> >> >false decision, but only the ellaboration of sets of options (based
> >> >on the agent's own knowledge), with selection being made on
> >> >a random basis or other criteria that are influenced by the
> >> >environment (this would be my choice). It would still be a
computational
> >> >implementation, with the results of such computations being
> >> >frequently unpredictable by the "programmer". In this regard, I can
> >> >accept that the "whole system" (made by the computational system
*plus*
> >> >its immediate environment) as being possibly non-computational.
>
> >> My primary concern is with how an agent acquires its own knowledge.
> >> I don't see much possibility for a computationalist system.
>
>
> >Acquisition of knowledge (or perhaps more appropriately, construction
> >of knowledge) is also among my main interests, and I so far haven't
> >found good enough reasons to think that a computational system
> >is incapable of building it (provided that this system is left to
> >interact with a natural environment).
>
> It is difficult to prove that a computational system couldn't
> build knowledge. One of the difficulties is an inadequate
> account of what is knowledge.

That's an important concept, and perhaps we should try to come
up with a "non philosophical" version of the word knowledge.

>
> Still, machine learning has not shown anything that approaches human
> learning. And it isn't for lack of trying. The best "computational"
> learning systems cheat. That is, they depend on reward systems
> (reinforcement systems) which are outside the computation. You might
> say that my current direction is to investigate autonomous reward
> systems.

I agree that machine learning, the way it is today, is a far cry from
what has to be done. But I believe that initial autonomous rewards
(and with initial I mean "primary" or infant-like) can be obtained
from the recognition of very salient statistical properties of the
sensory signals. These properties are insufficient for the development
of high-level cognition, but they seem to be important "first steps"
towards it.

>
> >I agree that knowledge creation is the key, but I don't think
> >we have to previously insert any sophisticated knowledge into the
> >agent. An animal may be born with a limited set of requirements
> >(need of food, need of physical comfort, etc) and from that (plus
> >a suitable computational brain) it may develop its own knowledge.
> >It doesn't seem necessary to insert knowledge into such an agent
> >other than some simple principles that allows it to grow its own.
>
> I don't have any problem with that. But notice how you have to leave
> computer systems and go to animal systems for good examples. So what
> is it that the animal has that computers don't have?

That's the million-dollar question (or better, its answer). I would
say that animals have "embedded" the ability to adapt their behavior
to the dynamics of the surrounding environment. This tends to make
me think about an adaptive dynamical system. My hypothesis is that
computers can be programmed to "simulate" the behavior of this
dynamical system.

>
> >> With a system based on pragmatic judgement, it can use trial and
> >> error to see what works. When its only standard of judgement is
> >> that defined by the programmer, I can see only quite weak ways
> >> of learning.
>
> >I also have some doubts that one have to have a sophisticated standard
> >of judgment in order to start building the knowledge. In other words,
> >I propose that it seems possible to find some simple computational
> >criteria which can be used as starting point of a simple organism
> >capable of such knowledge building. Yes, this standard will be
> >defined by the programmer, but by a programmer who can reason that
> >this standard is also the origin of knowledge in other biological
> >organisms (thus this standard is the result of theoretical
> >considerations). The standard I propose is statistical in nature
> >(and notice that I'm not following Bill Modlin's steps).
>
> With some simple programming, say code that evaluate [f(x)]^2 - x,
> a computer might be able judge the adequecy of a square root
> computation.
>
> A human, using that ice-cream tastes good, might get to discovering
> life forms on Mars.
>
> With computational standards, how do you get outside the narrow
> ranged determined by those standards?

I think I understand your question and it seems obvious that
we have to let the organism come up with its own standard. What
I propose is that the common idea behind all the standards of
all intelligent animals of the planet has something to do with
the discovery of the most relevant statistical properties of
stimuli.

Sergio Navega.

Relevant Pages

  • Animal Rights Myths FAQ v1.2 (1/2)
    ... of animals in biomedical research. ... A gauge of the biological similarity between animals and humans ... MYTH 2.2: ... standard pharmacology textbook describes chloroform as follows: ...
    (sci.answers)
  • Animal Rights Myths FAQ v1.2 (1/2)
    ... of animals in biomedical research. ... A gauge of the biological similarity between animals and humans ... MYTH 2.2: ... standard pharmacology textbook describes chloroform as follows: ...
    (sci.answers)
  • Animal Rights Myths FAQ v1.2 (1/2)
    ... of animals in biomedical research. ... A gauge of the biological similarity between animals and humans ... MYTH 2.2: ... standard pharmacology textbook describes chloroform as follows: ...
    (sci.answers)
  • Animal Rights Myths FAQ v1.2 (1/2)
    ... of animals in biomedical research. ... A gauge of the biological similarity between animals and humans ... MYTH 2.2: ... standard pharmacology textbook describes chloroform as follows: ...
    (sci.answers)
  • Animal Rights Myths FAQ v1.2 (1/2)
    ... of animals in biomedical research. ... A gauge of the biological similarity between animals and humans ... MYTH 2.2: ... standard pharmacology textbook describes chloroform as follows: ...
    (sci.answers)
Quantcast