6
$\begingroup$

For example, the prefix term 'micro-' in 'microfinance' has a different purpose from the 'micro-' in 'microeconomics'.

I heard all 'interpretations' of quantum mechanics give exactly the same answer to every measurement so they are all equally correct. Is that the same use of the term 'interpretations' as in 'interpretations' of probability?

Context:

  1. In 2011, learned mathematical (frequentist) statistics in 2011 as a quant undergrad.

  2. In 2014, I encountered the aforementioned.

Sheldon: Okay. Um, what is the correct interpretation of quantum mechanics?

Howard: Since every interpretation gives exactly the same answer to every measurement, they are all equally correct. However, I know you believe in the Many Worlds Interpretation, so I’ll say that. Now do you think I’m smart enough?

  1. In 2015, I discovered the Bayesian interpretation of probability as a quant grad eg Bayesian logit model - intuitive explanation? and that 99% of my statistics were frequentist.

So Bayesians and frequentists interpret probability differently leading to the things like Lindley's parardox, but they both follow Kolmogorov's axioms and Bayes' theorem so they will never differ on $\mathbb P(A)$ where $A$ is an event in $(\Omega, \mathscr F, \mathbb P)$

Is that the same idea as the use of the term 'interpretation' in quantum mechanics?

If no, why exactly?

If so, elaborate if you want.

$\endgroup$
11
  • 1
    $\begingroup$ This seems like more of a philosophy question than a physics one to me. $\endgroup$ – By Symmetry Dec 8 '16 at 17:22
  • 2
    $\begingroup$ Not all interpretations of quantum mechanics are experimentally equivalent. For example the objective collapse theory predicts that we should not be able to prepare coherent quantum states of certain size. What's more, some interpretations have nothing to say for certain domains of physics, for example Bohmian mechanics can not describe quantum field theory (at least as far as I understand). Finally, we have not exhausted the consequences of our interpretations, so we may be surprised to learn that different interpretations make different predictions about nature that we did not anticipate. $\endgroup$ – Joel Klassen Dec 8 '16 at 18:34
  • 2
    $\begingroup$ Whether interpretations of quantum mechanics are really just interpretations of probability depends on the interpretation, actually! $\endgroup$ – knzhou Sep 1 '20 at 0:10
  • 1
    $\begingroup$ Why do you mention the "Big Bang Theory" (2014)? Because of the word "chance"? $\endgroup$ – Deschele Schilder Sep 1 '20 at 0:46
  • 1
    $\begingroup$ Aha! I understand. Too bad the other questions were closed! Great fun, TBBT ;-) $\endgroup$ – Deschele Schilder Sep 1 '20 at 23:20
1
$\begingroup$

Interpretation in quantum mechanics is explaining how seemingly counter-intuitive quantum mechanical behavior comes about. There exist many interpretations.

Interpretations of probability are explaining what probability is (i.e., how it is correctly defined). There are exactly two of them: frequentist and Bayesian interpretations - I do not discuss them here, since I have done it just yesterday in an answer to a different question.

In other words, these are two different disciplines that require explaining/interpreting rather different things. (Moreover, QM is a physical discipline, whereas probability theory is more a mathematical one, although the need for interpretation comes from applying it to real data - not necessarily physics data.)

In terms of philosophical standpoints there is some overlap between the interpretations of QM and those of probability, as pointed out in the answer by @EricDavidKramer.

$\endgroup$
7
  • $\begingroup$ Part 1. Right so your last paragraph I think is not really relevant to what my question - which i recently edited to hopefully be clearer - was meant to be. Similarly, some of the other answers I think are not really relevant or maybe I don't really understand those other answers. $\endgroup$ – BCLC Aug 26 '20 at 11:43
  • 1
    $\begingroup$ Part 2. Anyway, thanks for answering. so you're saying that interpretations of quantum mechanics are explanations of physical (or mechanical or whatever) phenomena but while interpretations of probability are more philosophical (or even ontological or whatever) into the nature of what probability is? $\endgroup$ – BCLC Aug 26 '20 at 11:43
  • 1
    $\begingroup$ @BCLC Precisely! QM and probability are really two different animals, even though probability is essential for QM. $\endgroup$ – Roger Vadim Aug 26 '20 at 11:45
  • 1
    $\begingroup$ 3. My answer is similar to the second part of that by Guy Inchbald's. I don't quite agree with his view of Bayesian/frequentist interpretation and the following suggestions about the time reversal symmetry $\endgroup$ – Roger Vadim Aug 26 '20 at 11:52
  • 1
    $\begingroup$ 4. It is disputable. There is "shut up and calculate school", usually associated with Feynmann or Landau, which takes condescending view of interpreting QM as philosophical and irrelevant to practice. Others however believe that we may eventually come up with the experiments to test different QM interpretations, and then it is more than mere philosophy. $\endgroup$ – Roger Vadim Aug 26 '20 at 11:55
3
+50
$\begingroup$

There are similarities and differences.

In probability, both interpretations give the same $P(A)$ for an event. But the interpretation of the even is the same: the event is absolute, i.e. it really happened.

In quantum mechanics, all interpretations give the same probabilities for a measurement outcome but the interpretation of the measurement outcome itself may vary.

In collapse theories, the measurement outcome is absolute and really happens, but causality is violated.

In many worlds, there is no single measurement outcome. The world becomes a superposition two worlds, each with its own outcome, and the observer in each world seeing (subjectively) a single outcome.

In Quantum Bayesianism, the probabilities and the measurement outcomes are completely subjective. This is actually a reformulation of probability theory where probabilities can interfere like waves. (This is because, in QM, the probability for a measurement $a$ is given by the square of the probability amplitude, $p=|\psi(a)|^2$. Since $\psi$ is, in general, a complex number, we can get interference between different classical processes by taking superpositions $\psi = \psi_1+\psi_2$.)

A recent paper (referred to in the following article) makes some of these ideas rigorous: https://www.sciencemag.org/news/2020/08/quantum-paradox-points-shaky-foundations-reality

But the point here is that the different interpretations of QM predict a different final state of the world and of the system after a measurement is made. However, how that measurement outcome is perceived by the observer making the measurement will be the same in all interpretations. So there are no observable differences between the different interpretations, only inobservable differences.

$\endgroup$
9
  • 1
    $\begingroup$ I think you make a very good point. It is however not the probabilities, but the probability amplitudes that interfere - I mention it only because it seems that quite a few users here are not sure about the difference between probability theory and quantum mechanics, and whether the former is the same in quantum and classical physics. $\endgroup$ – Roger Vadim Aug 25 '20 at 12:04
  • 1
    $\begingroup$ That's a good point. I just added your point to my answer. I was in a hurry before and it seemed complicated to discuss so I left it out. Please tell me if you agree with my explanation. $\endgroup$ – Eric David Kramer Aug 25 '20 at 12:45
  • 1
    $\begingroup$ Your "recent paper" was published in 1939. How old are you? :) $\endgroup$ – D. Halsey Aug 25 '20 at 13:27
  • 1
    $\begingroup$ Oops wrong paper! Changing the link right now. I'm 33. $\endgroup$ – Eric David Kramer Aug 25 '20 at 13:35
  • 1
    $\begingroup$ Thanks. That's a simpler question. Will update my answer. $\endgroup$ – Eric David Kramer Aug 26 '20 at 13:14
0
$\begingroup$

It has been said that where conventional (frequentist) probability looks forward and asks what probably will happen, Bayesian probability looks back and asks what probably caused something to happen.

This can be seen as reflecting time-reversal symmetry in quantum theory, they are not different interpretations of the theory.

Interpretations of quantum physics are concerned with the underlying physical mechanism; does God play dice with us or is there some hidden variable we have not figured out? What goes on in between measurements? Are quantum interactions transactional? Does reality constantly split into multiple parallel universes (yes, really, some physicists take this seriously!)? and so on.

It is probably (sic) correct to say that interpretations of probability play into some interpretations of QM.

$\endgroup$
3
  • $\begingroup$ thanks Guy Inchbald! is your answer similar to Vadim's? i actually notice that i think your answer and Vadim's are the only ones to understand what i meant to ask. i guess my question was kinda unclear. i edited it recently. $\endgroup$ – BCLC Aug 26 '20 at 11:44
  • 1
    $\begingroup$ Yes, I think we are saying much the same thing. Some interpretations of QM employ frequentist probability, others employ Bayesian probability, others probably haven't thought about that bit. $\endgroup$ – Guy Inchbald Aug 26 '20 at 12:36
  • $\begingroup$ Thanks! PS dear downvoter, why? $\endgroup$ – BCLC Aug 31 '20 at 23:47
0
$\begingroup$

Bayesianism often seems to be regarded as a modern interpretation of probability theory, which displaced frequentism, but actually it was the way in which probability theory was always understood except for a short period in the first part of the C20th. I think it was well summarised by Maxwell:

“The actual science of logic is conversant at present only with things either certain, impossible, or entirely doubtful, none of which (fortunately) we have to reason on. Therefore the true logic for this world is the calculus of Probabilities, which takes account of the magnitude of the probability which is, or ought to be, in a reasonable man’s mind.” James Clerk Maxwell (ed. P. M. Harman), 1990, The Scientific Letters and Papers of James Clerk Maxwell, Vol. 1, 1846-1862, p.197, Cambridge University Press

Frequentism is then seen as a particular way in which probabilities can be assigned in a reasonable man's mind. IOW Bayesianism and Frequentism are not opposing interpretations, but rather Bayesianism is a general interpretation and Frequentism amounts to a special case, which does have a great deal of practical application.

I would contrast this with so-called interpretations of quantum mechanics. These are just waffle. Even von Neumann's quantum logic is so mangled by Wikipedia that it has no bearing on what quantum logic actually is.

The predictions of quantum mechanics are not derived from interpretation, but from the mathematical structure defined by the Dirac–von Neumann axioms. Interpretations, by and large, consist of hand wavy arguments which bear very little relation to the axioms. In some cases, such as Bohmian mechanics, they ignore theorems following from the axioms (which prohibit determinism). In others, such as many worlds, they bear no relation to the axioms, and discuss things like "the wave function of the universe" which is a nonsensical concept in axiomatic quantum mechanics.

Many interpretations, particularly information theoretic interpretions, and interpretations like quantum Bayesianism, and Saul Youssef's "exotic probability theory" simply regurgitate in different words what was already said by Dirac and von Neumann, without adding anything of interest and usually without making direct reference to mathematical structure.

The exception is the orthodox, or Dirac–von Neumann interpretation, in which qm is understood as a mathematical structure for the calculation of probabilities of outcomes of measurement (either Bayesian or frequentist probability can be used). The difference in mathematical structure between this and classical probability theory is precisely that quantum mechanics does not admit hidden variables, whereas in classical probability theory it is often assumed that results are determined by unknowns. The reasons for this mathematical structure are the subject of theorems, not interpretation. I have written much more on this in my books (see profile) and in e.g. The Hilbert space of conditional clauses

$\endgroup$
5
  • $\begingroup$ Edited to be clearer. My question is about the use of the term 'interpretation'. Thanks for answering. Is your answer the same? $\endgroup$ – BCLC Aug 26 '20 at 11:39
  • $\begingroup$ anyhoo i'll just upvote your answer for effort since it was my fault for being unclear $\endgroup$ – BCLC Aug 26 '20 at 11:57
  • 1
    $\begingroup$ Yes, my answer is the same. I meant to clarify what I mean by interpretation, using the example you gave of classical probability. There is only one "interpretation" of qm which I regard as being an interpretation in this sense. The others are not really interpretations at all imv, just hand waving unrelated to mathematical structure. $\endgroup$ – Charles Francis Aug 26 '20 at 14:45
  • $\begingroup$ Thanks! PS dear downvoter, why? $\endgroup$ – BCLC Aug 31 '20 at 23:47
  • $\begingroup$ There is only one "interpretation" of QM which I regard as being an interpretation in this sense That's your interpretation! $\endgroup$ – Deschele Schilder Sep 2 '20 at 0:23
0
$\begingroup$

The non-probabilistic interpretation
The non-probabilistic interpretation of quantum mechanics proposes a mechanism underneath the surface that leads to a probabilistic interpretation of QM on the surface. The Bohmian interpretation is an example: underneath the surface lie hidden variables which lead to a probabilistic theory, but one that is explainable (in contrast to a truly probabilistic theory). Compare it to the unpredictable Brownian motion in a fluid, where the fluid represents the hidden variables (continuous or discrete around very small distances, say the Planck length) Non-local Hidden Variables interpretations (like Bohmian QM) automatically are non-probabilistic, since an underlying deterministic mechanism is incorporated in this theory/interpretation.

The probabilistic interpretation
This interpretation states that QM is inherent, non-reducible probabilistic. The question remains: Why is this so? There is no answer to this question in this interpretation. It was just stated for a fact (by Born).

Presently, most interpretations are probabilistic in their approach. The question then remains: How can anything be fundamentally probabilistic? The Bohmian interpretation gets rid of this unexplainable interpretation (which was just stated to be probabilistic by Born to explain the outcome of observations; it could, however just as well have been that today we would be talking about a non-probabilistic interpretation, which seems to make more sense to me).

$\endgroup$
5
  • $\begingroup$ Edited to be clearer. My question is about the use of the term 'interpretation'. Thanks for answering. Is your answer the same? $\endgroup$ – BCLC Aug 26 '20 at 11:39
  • $\begingroup$ anyhoo i'll just upvote your answer for effort since it was my fault for being unclear $\endgroup$ – BCLC Aug 26 '20 at 11:57
  • 1
    $\begingroup$ My answer is about interpretations too. There are basically two interpretations: the realistic and the subjective. $\endgroup$ – Deschele Schilder Aug 26 '20 at 17:19
  • 1
    $\begingroup$ Why the downvote? $\endgroup$ – Deschele Schilder Aug 30 '20 at 11:43
  • $\begingroup$ Thanks! PS dear downvoter, why? $\endgroup$ – BCLC Aug 31 '20 at 23:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.