On 11 Dec 2015, at 02:46, Dan wrote:
Digital physics rests upon the Strong Church-Turing Thesis, which
posits that nature does not admit non-computable real numbers:
If digital physics is correct, then my body is Turing emulable. But
then, the universal dovetailer argument shows that you need to derive
physics from a statistics on the natural number computable relations,
which entails that physics cannot be computable. This means that
digital physics entails its own contrary, and thus digital physics is
self-contradictory.
This does not prevent digital physics from having applications, but it
is wrong on the fundamental level.
https://en.wikipedia.org/wiki/Church
%E2%80%93Turing_thesis#Philosophical_implications
AI researcher Ben Goertzel discusses the "Hypercomputable Humanlike
Intelligence hypothesis, which suggests that the crux of humanlike
intelligence is some sort of mental manipulation of uncomputable
entities – i.e., some process of “hypercomputation” [1-6].":
http://goertzel.org/papers/CognitiveInformaticsHypercomputationPaper.pdf
This recent article seems to imply that there non-computable real
numbers exist in nature. Would you agree? If so, does this not
falsify the Strong Church Turing Thesis? That then seems to make the
HHI hypothesis possible, and if the HHI hypothesis could be verified
then according to Goertzel's argument science would never be able to
describe cognition. (He's concerned about the implications that
would have for neuroscience and AI, naturally.):
http://factor-tech.com/connected-world/21062-a-fundamental-quantum-physics-problem-has-been-proved-unsolvable/
I've only come across this mailing list after reading Standish's
"Theory of Nothing." On my blog, I have been compiling various
recent work regarding the fabric of reality. It attempts to tie
together Chaitin's metabiology (life as evolving software),
Hoffman's Interface Theory of Perception and Dynamics of Two
Conscious Agents, modeling by Deutsch/Wallace of the Born rule using
rational agents and decision theory, Standish's derivation of QM as
implementation of evolution, Alex Wissner-Gross's Entropic Causal
Forces relating intelligence to emergence/entropy, Frieden's Extreme
Physical Information which seems to explain evolution as a program
applying the anthropic principle and operating on particles and
ideas (which are simply dual aspects of information, taken from a
multiverse of infinite bit strings a la Standish or plentitude of
mathematical structures a la Tegmark). It also considers the work of
Bolognesi in "Algorithmic Causal Sets for a Computational Spacetime"
and Schmidhuber's Goedel Machine,Theory of Creativity, and universal
dovetailer. I suppose my current hypothesis about the fabric of
reality (prior to fully grasping Tegmark's hypothesis) can be
summarized as follows:
Your attribution are wrong. See my papers in my URL below, or on
research gate for the more recent one. But Tegmark and Schmidhuber
missed the first person indeterminacy, and mainly the global first
person indeterminacy, which is that if we are machine we are
duplicable, and this leads to a first person indterminacy. See perhaps
this paper:
http://iridia.ulb.ac.be/~marchal/publications/SANE2004MARCHALAbstract.html
Then they miss the fact that all computations are emulated in virtue
of the true relation between numbers (or of the true relations in any
Turing complete theory), which means that to make any prediction on
first person experience (which is what we do when doing an experience
in physics) we need to make a statistics on all computations seen from
the first person point of view (the global FPI). Nature is reduced to
an emerging first person (sharable) reality.
Particles and ideas are two sides of the same coin: information.
Information is a pure third person construct, so it cannot help to
solve the problem. You need to structure information by the first
person apprehension. See the paper mentioned above to understand this
with more details. I explain this on this list since some time. You
might consult the archive or read the sane04 paper.
Driven by the anthropic Extreme Physical Information principle that
nature is kind to observers (as efficiently as possible-- Occam's
razor), ideas and "particles" evolve algorithmically in much the
same way that genes do. What remains to be answered is why nature
should be kind to observers, or - perhaps more fundamentally- what
does it really mean to be an observer? Is it just a rational agent,
similar perhaps to a Goedel machine? When we say nature is kind to
observers, do we really imply that the rational agents ensure their
own existence by maximizing novel pattern compression progress and
possible future actions (as suggested by Hoffman's Conscious Agent
Thesis)? Please forgive me if I repeat myself or produce "word
salad" -- it can be difficult for me to sequence my thoughts
linearly as I struggle to get them all down, so upon request I will
clarify what I mean by anything above. I appreciate your patience
and am eager to read your reply.
I suggest you read the paper above:
http://iridia.ulb.ac.be/~marchal/publications/SANE2004MARCHALAbstract.html
So that you can catch the wagon where we have arrived. My work on the
"Goedel machine" (which I call Löbian machine), the universal
dovetailer argument are much older, and much advanced that most
citation you give. The only problem is that in the math part I use
some amount of mathematical logic which is not very well known. Your
intution is right: ideas evolve, but not like algorithm, which would
single out one basic algorithm at the origin of the physical reality.
That is possible, but must be derived from the statistics on all
computations. It makes physics not entirely computable, but perhaps no
more complex that computable + a random oracle (to use the FPI).
Bruno
--
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 http://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 http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
