Fundamental Compressionist Philosophy.

[Andrew Stanworth]

Gerry wrote;-

| A convention has grown up that the kind of computing done by conventional
| computers is 'symbolic' (because they use letters, numbers etc) and the
kind
| of computing done by a neural net is 'sub-symbolic', 'non-symbolic' or
| 'connectionist' because (artificial) neural nets (and perhaps natural ones
| too) don't seem to use the kinds of chunky symbols that ordinary computers
| use. But most neural nets are actually run on ordinary digital computers
| (which use the symbols '0' and '1' at their basic level). So the
distinction
| is not a very precise one.

I get what you mean - it is a distinction between what can be considered to
be the sub-units of a language as opposed to the sub-units of a process.  I
understand that much of academic computing, especially artificial
intelligence, has come from a linguistic standpoint (which I feel is too
restrictive a landscape - but that's my view), consequently it seems
reasonable to view language elements are carrying a symbolic 'meaning'
(i.e., a public standard/consensus) whereas neural nets use a private set of
symbolic relationships.  Maybe it would be better to describe these
relationships as 'publicly symbolic' and 'privately symbolic'?  Does my
definition seem more precise to you?

| When I said the ICMAUS ideas were intended to be relatively abstract, it
was
| this rough-and-ready distinction that I had in mind. I was really trying
to
| respond to your suggestion that "your system belongs in the 'neural
network'
| family of architectures." The framework uses symbols, and current
| implementations run on an ordinary digital computer (which uses binary
| symbols). So in that sense, the framework is symbolic. But the underlying
| concepts (information compression by matching and unification of patterns)
| are not, in themselves, symbolic (you may dispute this). And it seems
| entirely possible to me that the whole framework could be implemented
using
| neural mechanisms (which, according the rough-and-ready distinction) are
not
| symbolic. In short, I was trying to say that the system is not
intrinsically
| 'neural' and is not fundamentally symbolic either (in the rough sense of
| that term).

I see where you are coming from now, and see no problem with it viewed from
your perspective.  I was coming more from the "if it looks like a duck,
walks like a duck, and quacks like a duck - it's a duck" standpoint, which
accepts no functional level below the symbolic in its pure sense.  It seems
we may be having purely semantic problems, rather than substantive ones.
Thank you for acquainting me with an unfamiliar convention.

| "I don't understand what you mean by more abstract, since 'symbolic' is as
| abstract as I can conceive". Are analogue computers (using things like
| continuously-varying voltages) symbolic? If not, then the features that
are
| shared by digital and analogue computers are, presumably, more abstract
than
| 'symbolic'.

I don't accept that it is possible to 'know' any 'true' analogue entity (or
for such a thing to exist in sensory experience).  I believe that our
conceptions of the 'continuous' are entirely synthetic.  Leaving this stance
aside, should the symbol be considered to be the fundamental unit of
communication (as I do indeed consider it) - any analogue computer whose
varying-voltages were not to be considered symbolic, would never be able to
be communicated with, nor communicate anything to us.  Hence it could not
compute (in our experience), since it would be devoid of tangible input or
output.  Consequently, I still see nothing more abstract than the symbolic
(in a purist sense).  In other words, my definition of 'symbolic' is a
packet of abstracted experience.  In other words, to me, 'abstract' and
'symbol' relate to the same entity, consequently one can not exist at a
level fundamentally below that of the other.  It is because of this, and
other concepts, that I had difficulty relating to your use of the term
'symbolic' and the concept of more abstract than symbolic.  Thanks again for
clarifying your position and thanks also for posing an interesting question.

Best wishes,

Andrew.