On 27 Dec 2017, at 21:56, Brent Meeker wrote:
On 12/27/2017 12:44 PM, Bruno Marchal wrote:
On 27 Dec 2017, at 17:56, John Clark wrote:
Computationalism is the idea that the brain is an
information processing system and that a computer
can perform all the complex behaviors that would be called
intelligent if it were done by a human;
That is not computationalism. That is the weak AI thesis.
The strong AI thesis is that such machine would be conscious, and
computationalism is the even stronger assumption that "I am
machine" and that I would survive in the clinical usual mundane 1p
sense with a copy made at some right substitution level. But then I
get undetermined by the fact that no digital machine can ever
determine which machines she is,
Suppose she is provided with the design drawings etc used to build
her. Then she can point to them and say, "That is what machine I
am." And she could even write an emulation of herself to predict
what she would think.
Very good. That this is possible (consistent) and arithmetically
necessary, is given by the second recursion of Kleene. Its
formalization (translation in the language understandable by the
machine) is given by Gödel diagonalization lemma. It is the base of
theoretical computer science.
This is used in the theoretical inductive inference theory, like in
the work of Case and Smith. There are interesting tradeoff hierarchies.
Now, that is only what I called third person self-reference, and the
arithmetical version of the first person is a non arithmetical
predicate, yet "known" by the "correct" machine/number. That "things"
is in relation with the non arithmetical notion of arithmetical truth.
It is related with the least non constructive ordinal, omega_1^CK (CK
for Church and Kleene).
The universal machine lives, and "crash" in between the computable and
the non computable.
Bruno
Brent
nor which computations support her in arithmetic. But relative
measure makes still sense, and would justify the appearance of the
physical laws from the arithmetical internal pov.
Note that I use the term computationalism also in a weaker sense:
as the "brain" is any portion of the physical universe which could
be needed for this behavior+consciousness.
That nuance between weak AI/ and computationalism is of course
quite important to understand the mind-body problem, and its
embryonic computationalist solution. Obviously Weak AI thesis
cannot handle the hard problem (which is the mind-body problem).
computationalism does NOT insist that everything is information
processing
It is not part of the assumption. Indeed.
(although it does not rule out that possibility).
Actually, computationalism implies it, (but you need to grasp UDA
step 3 and sequel to get this).
Then without step 3, you can still use just the weak AI thesis to
get that the machine will behave like understanding that physics is
derivable from a consciousness theory, but logically, you are free
to make them into zombie. Again, such nuances are important in
philosophy of mind, metaphysics, etc.
If a n AI is made that is as smart or smarter
than humans then we'll know it's right,
That is the weak-AI thesis. But we will evaluate relative
competence only.
if that proved to be impossible then we'll know
computationalism is wrong.
OK.
That would be the case if the material hypostases get quite
different from the logic of the observable inferred from observation.
Most of the things on your list, although very interesting in
themselves, are irrelevant as far as the truth or falsehood
of computationalism is concerned. For example,
maybe we can't make perfect predictions of what physical systems
will do because an infinite number o f calculations would be
needed and that would be impossible, or maybe a finite but
astronomically large number of calculations would be needed and
that would be impractical, or maybe calculations have nothing to
do with it and some effects have no cause and true randomness
exists; it doesn't matter because for whatever reason the
fact remains that the human mind can't predict with perfect
precision what a physical system will do.
And computationalism has a lot to say about
intelligent behavior but it has nothing to say about
consciousness, no scientific theory does because we all
only have direct access to one conscious being and good
science can't be done with just one data point , and
any argument in support of the proposition that a computer
that behaves intelligently is not conscious could also be used in
support of the proposition that none of our fellow
intelligent behaving human beings are conscious.
I am a machine, in the strong computationalist sense, entails that
neither the physical reality, nor the bio-psycho-theo-logical
reality can be 100% computable.
So, computationalism is different from digital physics (the belief
that the physical universe is itself a computable object). I have
given reasons that digital physics is inconsistent (with or withput
computationalism).
The non computability of the individual prediction in QM confirms
the existence of the global First Person Indeterminacy on
infinities of computations in arithmetic.
Bruno
John K Clark
============================================================
On Wed, Dec 27, 2017 at 2:18 AM, Jason Resch
<[email protected]> wrote:
Bruno has often spoken of the confirmations of computationalism
known to date, and of the idea that it has passed many tests and
not been falsified so far. I was hoping with this post to gather
a complete list of those tests. What things in physics would
disprove computationalism, and what tests has it passed so far?
Below I try to collect a complete list from memory but it may be
faulty. I ask that others might add to this list or correct
things I have gotten wrong:
Tests and statuses of each test:
Non-emulability of physical laws
Non-discreteness (continuousness) of space time --- (somewhat
confirmed)
Infinite computation needed for tiniest amount of space ---
(mostly confirmed)
Quantum Mechanics
Uncertainty principal (inability to collect exact and complete
knowledge about environment) --- (confirmed)
Indeterminancy --- (appearance of randomness is confirmed,
explanation for being an "appearance only" i.e. first person
indeterminancy vs. fundamental randomness is made is plausible)
Born Rule?
Quantization of Energy?
Unitarity
General Physics (I am not sure if these are required by
computationalism, and could use some more help on these)
Linearity of physical laws?
Time reversibility?
Conservation of Information? (e.g. black hole information paradox)
Finite Description of Quantum States (e.g. Bekenstein Bound)
Link between Entropy and Information (e.g. Landauer's Principle)
Existence of a "Time" dimension?
Consciousness
Qualia - The non-communicable nature of some observations ?
Finiteness - (finite memory / age / information content of
experience)?
Are there other things I am missing? If any of the items I have
included are incorrect I would greatly appreciate any correction
and further insight.
Perhaps most interesting are any predictions which are presently
unconfirmed, as this would lead to predictions which could later
be tested and lead to a refutation of computationalism (or if
passed, yield further evidence for computationalism).
Jason
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected].
To post to this group, send email to [email protected]
.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected].
To post to this group, send email to everything-
[email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
