On 9/9/2011 1:37 PM, Evgenii Rudnyi wrote:
On 09.09.2011 21:58 meekerdb said the following:
On 9/9/2011 11:35 AM, Evgenii Rudnyi wrote:
On 06.09.2011 22:25 meekerdb said the following:
On 9/6/2011 12:43 PM, Evgenii Rudnyi wrote:
I was talking about realism in a sense that universals exist (I
am not sure if this could be generalized for all things). My
first naive/crazy idea was that this could give some basis to
produce qualia related to notation. Neurons somehow distill
universals from things and report them.
On the other hand, if we are to write a program that should
classify objects, then this program should have some
dictionary with categories. That dictionary in some sense
should exist.
Wouldn't those neural net face recognition programs be an example
of this. They start out not knowing anyone's face. But then with
training they learn to recognize Brent and distinguish him from
Evgenii. Each instance of the Brent image is a little different
from the other instances but it assigned the same classification
for purposes of access or other action. In effect it has invented
"Brent" and "Evgenii" as universals. The 'dictionary' then exists
as the combined information of the program and memory. The
persistent patterns in memory are analogous to dictionary
entries. The imaging and actions provide the meaning of these
entries.
I like more to take an example with a human being rather than with
a name, so let me consider a term "a human being". So, after all a
neural net is some map. It takes some visual, audio, tactile, etc.
inputs, processes them and produces some token. What happens then?
Presumably it puts this token to the dictionary that produces
qualia for the homunculus in the brain (or whomever, this does not
matter at this point). Now I would say that if that final qualia
corresponded to "a human being" is the same in all brains, than
this is realism. If different, then this is nominalism.
Evgenii
I don't think that's the distinction between realism and nominalism
in their theory of universals. It's my understanding that the realist
says that there really are human beings in an objective sense (where
"objective" may really just refer to intersubjective agreement).
While the nominalist says "human being" is just name we give to a
category created arbitrarily and we could just as well have defined
it as hairless bipeds and include ostriches and shaved kangaroos.
Brent
Yes, you are right. My interpretation is different from the conventional difference
between realism and nominalism. Here one says indeed that each person has something that
exists in the objective sense and this something is "a human being". Well, it we treat
qualia ontologically, then I guess, this will be close to realism. Yet one can imagine
different scenarios. Under a conventional definition, qualia "human being" is tied with
a physical person in the classical sense of the realism. It is necessary however then to
explain how a homunculus in the brain retrieves that qualia from a physical person
(quantum consciousness?).
I think it is a category error to think of a token being put in a dictionary as evoking
qualia. I think qualia supervene on the conscious formation (and recall) of symbolic
(mostly language) narration which is put into memory (although possibly only short term).
In the neural net analogy, the perception of a person activates some part of the network
so that some word, e.g. "Bob", gets inserted in the stream of consciousness that it is
going into memory. "Bob" is retrieved only in the sense that some part of the network is
activated. There is no homunculus.
Brent
The scenario that I have described is different in a sense that the communication takes
place through physical processes that we know but at the end we may still think of
qualia in the ontological sense. Hence one could probably state that this is also the
realism (but definitely in some unconventional sense).
Evgenii
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
