home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.physics:21113 sci.math:16856
- Path: sparky!uunet!elroy.jpl.nasa.gov!sdd.hp.com!think.com!news!columbus
- From: columbus@strident.think.com (Michael Weiss)
- Newsgroups: sci.physics,sci.math
- Subject: Re: Bayes' theorem and QM
- Date: 14 Dec 92 09:56:16
- Organization: Thinking Machines Corporation, Cambridge MA, USA
- Lines: 47
- Message-ID: <COLUMBUS.92Dec14095616@strident.think.com>
- References: <1992Dec10.192451.9924@lmpsbbs.comm.mot.com>
- <1992Dec11.220139.14066@newshost.lanl.gov>
- <1992Dec13.204233.10835@galois.mit.edu>
- <1992Dec14.003452.12884@galois.mit.edu>
- <SRCTRAN.92Dec13221623@world.std.com>
- NNTP-Posting-Host: strident.think.com
- In-reply-to: srctran@world.std.com's message of 14 Dec 92 03:16:23 GMT
-
- In article <SRCTRAN.92Dec13221623@world.std.com> srctran@world.std.com
- (Gregory Aharonian) writes:
-
- Speaking of Bayes Theorem, one theorem I never see mentioned in many of
- these discussions, and I am almost afraid to mention it (since it can
- spawn many pseudo theories) is a theorem due to Gregory Chaitin at IBM,
- who proves that if you accept the basic properties of Integers, then it
- follows (in a very pretty argument) that statements about reality are
- all probabilistic, that any theory that uses the integers implicitly
- is bringing in randomness (or that's how I interpret his work).
-
- One could argue that some of the probability that comes into play with
- quantum theories comes from the fact that physicists use integers to
- experiment and predict, that there is an unstated axiom in physics that
- things are probabilistic.
-
- I know I am mishandling his ideas greatly, and would appreciate it if
- some of the theorists could read his work and see if there is any relevance
- to the axioms of physics. If nothing else, his papers are great reading.
-
- You are indeed "mishandling his ideas", more or less like saying "Goedel
- proved no theory can be complete", or "Heisenberg showed everything is
- uncertain".
-
- Chaitin showed that there is a real number Omega between 0 and 1 whose
- binary expansion satisfies two properties:
-
- (1) It can be defined arithmetically. That is, the proposition "the n-th
- bit of Omega's binary expansion is 1" is equivalent to saying "G(n) is
- true for the integers", where G is a formula with one free variable in
- the language of Peano arithmetic.
-
- (2) Omega is Solomonoff-Kolmogorof-Chaitin random. The definition of this
- notion is somewhat technical. VERY ROUGHLY speaking, Omega will pass any
- sequence of statistical tests that can be posed by a Turing machine.
- However, there are plenty of pitfalls on the path from this intuitive
- notion to a precise mathematical concept. (Consult Chaitin's book, or
- the entry in the Handbook of Computer Science on algorithmic
- information theory (I forget the authors). I also have notes in LaTeX
- that people are welcome to.)
-
- I agree it would be nice if a close connection could be forged between
- Chaitin's ideas and the foundations of QM. At the moment though such ideas
- are pure speculation; no one (to my knowledge) has shown how this can be
- done.
-
-
-