Quantum Bayesian Networks

April 29, 2011

Don’t Shut Up and Don’t Calculate

Filed under: Uncategorized — rrtucci @ 10:03 pm

Those practically-minded, results-oriented physicists who prefer spending their time working out the predictions of quantum mechanics (QM) and comparing those predictions to laboratory data, instead of rambling endlessly about the philosophical implications of QM, are often said to belong to the “shut up and calculate” school. It seems to me that the advocates of the multiverse “many-worlds interpretation” (MWI) of QM belong to the less desirable school of “don’t shut up and don’t calculate”.

MWI was first proposed in 1957 by Hugh Everett. More than half a century later, it still hasn’t spawned any useful results. And that, in my opinion, is the biggest flaw of MWI: that it’s not very useful. Is it possible to explain or predict or calculate any laboratory observations much more simply or elegantly using MWI than not using it? No. Like I said before, useless.

The vast majority of people working in quantum computing have never used MWI in their work. QM can easily stand without MWI. I can’t stress this fact enough. I hope that those new to the field of quantum computing don’t get turned off from the field because they get the mistaken impression that it’s based on MWI.

David Deutsch claims that MWI is necessary in order to “explain” a quantum computer. He has no mathematical proof of this claim. He argues that the claim must be true because of philosophy (his philosophy). Yawn. He even claims that he has devised a test that singles out MWI as the one and only possible interpretation of QM. His test requires a bizarre self-aware quantum computer artificial intelligence. Nothing like proposing an impossible-to-do, cryptic experiment to prove your point.

Advertisements

6 Comments »

  1. It is probably useless to make arguments here since you have probably heard of them before. But I will try..

    Situation 1: Before Quantum Computers.
    We had two mathematical formulations of QM. One was the classical one and the other one was the one by Everett. The difference was that Everett had gotten rid of one axiom – the axiom that claimed that wavefunction collapsed on a measurement of a system. Now, in every respect as far as predictions went there was no difference between the two formulations. But as numerous scientists have argued over the ages, the role of scientific theories is not to predict the result of experiments. Their role is to explain the universe around us. The prediction is a test that allows us to find out which theories to believe in. But sometimes it so happens that two different theories make the same predictions and then this test fails. We then have the second test – Occam’s Razor. The simpler theory wins. Consequently, as the proponents of the MWI have argued, MWI is the simpler theory because of one less axiom and so is the one we should take as our explanation of the universe.

    You have to realize that just because MWI came later in time doesn’t mean that it is worse theory. If orthodox QM had come later, it would have stood in the same position MWI stands today. People would have laughed (incorrectly) at it for not bringing anything new to the table. Yet they do. The two theories give different explanations of the universe around us.

    Situation 2: After Quantum Mechanics
    Deutsch has shown that there is an experiment for which the two theories make different predictions. The test of a large quantum computer implementing Shor’s algorithm or the like. Hence, we make our quantum computer and test. We will find out.

    Comment by Abdullah Khalid — April 30, 2011 @ 5:41 pm

  2. Hi Khalid, I don’t see why you think MWI gives better or simpler explanations. MWI is not NECESSARY for explaining Shor’s algorithm. We’ll just have to agree to disagree.

    Comment by rrtucci — April 30, 2011 @ 6:36 pm

  3. Deutsch has given an explanation for Shor’s in MWI. I have yet to see any explanation within the orthodox framework. There needs to be some sort of explanation before we can agree or disagree about which one is simpler.

    Comment by Abdullah Khalid — May 1, 2011 @ 6:06 pm

  4. This is kind of funny and supports my contention that the vast majority of people working in quantum computing do not use MWI. Someone called Blake Stacey posted in Scott Aaronson’s blog the following two comments:

    Blake Stacey Says:
    Comment #33 May 3rd, 2011 at 3:35 pm

    Chris Fuchs once told the story that he went around asking people who had made notable discoveries in quantum computation whether they had been inspired by Everett’s interpretation of quantum mechanics. One — I think it was Simon — wrote back, “Who is Everett and what is his ‘interpretation’?”

    Blake Stacey Says:
    Comment #36 May 3rd, 2011 at 3:55 pm

    OK, having said that, I had to go look. I think I was recalling it from PIRSA:07090068, where he quotes Jozsa as saying,

    I’ve known of Everett interpretation since the mid 1970′s and never really adopted/liked it, even from outset. It always was (and still is) a very vague and incomplete framework to me. … I’m not aware that the Everett ideas have ever played any significant role in my thinking on quantum things. I don’t have a clear impression of any particular imagery that I could name, underlying or guiding my quantum thoughts … I do not see that any quantum comp/info developments particularly support the Everett view in any way compared to any other prospective interpretations.

    Shor said that he knew of Everett before starting work on his factoring algorithm, but that “the idea was really more to use periodicity, and inspired by Simon’s algorithm.” To which the punchline is Simon saying, “Who’s Everett, and what’s his interpretation?”

    Comment by rrtucci — May 4, 2011 @ 1:05 am

  5. The MWI picture is getting elegant and natural once we recognize that it is not universe splits apart (it’s always in superposition state) but observer is constantly branching. Our memory perceives a single path from root to current node. In fact the “Time” from this point of view is not one dimensional – it seems one-dimensional only thanks to our perception.

    Comment by Stefanbanev — June 5, 2011 @ 4:17 pm

  6. “Deutsch has given an explanation for Shor’s in MWI. I have yet to see any explanation within the orthodox framework. There needs to be some sort of explanation before we can agree or disagree about which one is simpler.”

    This is exactly the point. The MWI provides an physical “explanation” for how the calculations in a QC take place. The other intrpretations do not do that. The only thing I ever hear back when I raise this point is the the QM formalism (which the MWI adopts) “predicts” that a QC will work. Of course it does. But prediction is not explanation, and what proponents of non-MWI are left with is a black box where the calculatons take place, but with no explanation for how and why they occur.

    Comment by Mike — June 15, 2011 @ 6:44 pm


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: