Re: consciousness, was Re: etc.


On May 28, 3:55 am, c...@xxxxxxxx (Curt Welch) wrote:

JGCASEY <jgkjca...@xxxxxxxxxxxx> wrote:
On May 27, 6:10 am, c...@xxxxxxxx (Curt Welch) wrote:

[...]


Do you honestly have no clue why it works so well even
after Michael explained it to you?


But clearly Michael has not explained it to him otherwise
he would understand. What constitutes an explanation? Is
it not that it changes the behavior of the one to whom
the explanation is directed?


And you might ask yourself why I wrote the comment above.
Was it because I thought he didn't read Michael's answer?
Or didn't understand it? It was because I felt he wrote
off the answer without giving it a reasonable amount of
thought. My prime purpose was to let him know that others
in the group felt that Michael had answered the question,
and that if he didn't understand the answer, he should
spend at least a little more time considering what had
been said.


The group may have felt that Michael answered the question.
But if the individual didn't understand the answer it was
not an answer for that individual. MO may as well have
been writing in a foreign language.



[...]

If you do not understand how to translate an example such
as the one you gave, into the language of behaviorism,
that only shows that you don't understand the language of
behaviorism - or even some of it's most basic concepts.


Just as an explanation in mathematical form my not be
understood because the person does not understand the
language of mathematics or even some of its most basic
concepts.


True, that is an issue.


This is why I asked the question what does "conditioning"
mean, so that I could perhaps understand (be changed) by
the 'explanations'. However no meaning to the word was
given except it was all very complex. In which case why
waste time talking to people who don't understand the
words you use if you are not prepared to explain them?


The basics of conditioning are simple and trivial.


According to the others it is not trivial.


It's taught in introduction to psychology classes. You
can find many web sites to explain the basics of operant
and classical conditioning and they will explain all the
language you need to understand. It's something anyone
that is actually interested in the subject should be able
to understand in a day of work.

My book shelf contains books on psychology and one spends
much of the book on the subject probably because it is an
old book I picked up at a second hand book shop. It even
has a chapter on your beloved reinforcement and other
terms used by the behaviorists of this group.


What don't you understand about operant and classical
conditioning?


I don't know. Just I apparently know jack *** about it
according to MO. It seems about as nebulous as the word
"consciousness" when you try and get a simple reply.


However, a basic understand of the effects is just the
start. To fully understand what the EAB has uncovered
after 50 years of work will take more time. I don't
pretend to understand much of it or understand the
language. But I understand the basic position the
approach takes and what it has been able to show us
about ourselves (to a degree).


If you want to understand it all, you are going to have
to do years of work. It's not something that can be
explained to you in a Usenet post.


And according to your confession above not something you
understand much of either including the language. I think
you have your own version of "behaviorism".


So exactly what don't you understand, and exactly what
do you expect someone here to be able to do for you?
If you want to understand, go buy some books and do a
lot of reading and studying. Or take some classes.


If you want to understand the basics, just Google on
"conditioning" or "operant and classical conditioning".

I have done just that including articles on efforts to
use reinforcement learning. The most interesting articles
on behaviorism were on how to train your dog.

Your interest is in building a universal general purpose
learning machine. You can google on that subject as well.
That is also one of my interests. But you have to start
with working examples which you can understand in the
terms used to actually build the machines.

The feeling I get is that as soon as you get a machine to
do something intelligent the cry is; "that is not real
intelligence" and as soon as you get a machine to do
some simple kind of classical or operant conditioning
the cry is; "that is not real conditioning".

That may well be the case but you have to start somewhere
and although the behaviour of these machines may not be
really like an animals behavior they are behaviors never
the less and ones we fully understand the causes of. And
they may not be "real" intelligence but they are not
random either.

Nothing GS has ever written have I understood and certainly
have no idea how to implement any of it in software. However
others have written software that produces interesting
behaviors and that is what AI is about for me. I don't need
a behaviorist to tell me if my robot has learnt to do a
little dance as a result of my pressing the reward button
to its initial random movements. Maybe not how an animal
learns but it is a changed behavior that to me is a kind
of learning and one I fully understand.


--
JC

.

Loading