non-random quantum mechanics

[Metric]

I recently came across some interesting work that was done about 10 years ago, concerning the "measurement problem" of quantum mechanics. Basically, the measurement problem refers to the strange non-deterministic "quantum jumps" and "randomness" that occur when a measurement is made.

This work, however, sets all of quantum theory in a new and interesting light. In this formalism, there are no "jumps" -- pure quantum mechanical states evolve to pure quantum mechanical states, completely deterministically. Probabilities arise only when "ignoring" part of a system in a well-defined way. However, it is shown by the author that this process of "ignoring" part of a system is common to all measurements, and thus the probabilities that we have always dealt with by introducing a seperate (and rather unsatisfying) postulate of QM emerge naturally due to an unavoidable "ignorance of the state" that is inherent in every measurement.

Interestingly, this work has not become widely known and appreciated. I contacted someone I know who works in this field and inquired why this was the case, and his impression was that the formalism (which makes extensive use of information theoretic concepts) was somewhat dense, and the work preceeded the huge interest in quantum information theory which has arisen in the intervening years -- i.e. he was ahead of his time. The author has since gone on to other things, and nobody else has made much of an effort to advance this stuff.

For those of you who are fluent in quantum mechanics and have an interest in the philosophical implications of "randomness" I recommend this paper:

http://xxx.lanl.gov/abs/quant-ph/9605002

[Borodog]

Very interesting. Thanks for the tip. How did you come across this paper, by the way?

[Metric]

[ QUOTE ]
Very interesting. Thanks for the tip. How did you come across this paper, by the way?

[/ QUOTE ]
I am studying general properties of information in covariant theories where no good time variable may be available (theories like GR have this property). One problem that shows up is this notion of "wave function collapse," which happens "instantly" in the standard approaches to quantum theory -- but "instantly" is ambiguous if there is no good time variable sitting around to reference (one can easily construct examples where you get two different answers to the same question and the theory does not choose between them). So I talked to someone who was aware of this line of work in which no collapse takes place at all -- everything remains unitary -- and the appearance of collapse is only due to information theoretic properties of the measurement process. Hopefully, it will lead to a notion of measurement that makes sense even in the covariant framework.

[gumpzilla]

Looks like a pretty long paper. I intend to read it in the next couple of days, and perhaps I'll post some thoughts or questions here.

[cambraceres]

This is a little strange, I'll have to make time for your find. This could be a benign restructuring of the elucidation that brings about no true progress. Like when Feynman declared the pringiple of indeterminacy to be unnecessary when QED was properly applied.

There is a curious lack of these types of posts here

Cambraceres

[Metric]

[ QUOTE ]
This could be a benign restructuring of the elucidation that brings about no true progress.

[/ QUOTE ]
That is always possible -- in fact it's my default assumption when someone claims to have a new way of thinking about QM. However, in this case, the formalism may allow some concepts to extend to the covariant framework in a way that was not possible before (that's what I am hoping, anyway). I.E. it's not just philosophically nice -- it actually helps you get the right answer.

[PokerPadawan]

Some of you may find it interesting that Chris Adami, the second author, is now a well-known artificial life guy.