Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Breaking Bell's Inequality with Monte Carlo Simulations in Python

bytepawn.com

Post reply on HN
Print view View original

11–20 of 64 posts

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#11

Post by eigenket »

n4r9 wrote: post #6

This is closely related to my PhD. It was many years ago but if I remember rightly there is no need for the assumption of determinism - Bell Inequalities hold just as well for random local hidden variables. Simulating the correlations with computer programs is an interesting idea, partly because it challenges to those who still believe in a "local" reality to demonstrate Bell Inequality violations in distributed clas…

You know something has gone horrifically badly when a paper begins with

> This reply paper should be read as a continuation of my previous reply paper [1], which is a reply published in this journal to a previous critique of one of my papers

We're way too deep in replies now, and anyone who values their time should get out now.

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#12

Post by eigenket »

GistNoesis wrote: post #9

The Bell's Inequalities are a test for the capacity for inductive reasoning of the pupil. If the pupil succeed he is not to be admitted to join the ranks of quantum physicist. You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore y…

> You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore you conclude that the world is non-local.

If you do this you're doing a bad job at being a teacher.

The way the argument should go is you start with a list of assumptions (of which locality is one), derive Bell's inequality from them, and determine that as Bell's inequality seems to be false in real experiments at least one of your assumptions was wrong. Then you can talk about quantum mechanics and explain which of these assumption are broken in quantum mechanics. If you have time you can have fun talking about different interpretations of quantum mechanics because (e.g.) Everettian Many Worlds is completely local, but still produces predictions matching quantum mechanics (and therefore breaks Bell's inequality).

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#13

Post by jahnu »

n4r9 wrote: post #7
danwills wrote: post #5

"a talented college physics student can do it" I'm afraid I don't qualify for being able to do that, but I feel like I'm tantalizingly close to understanding this overall - but I'm finding it hard to understand why the lower-right "TT" quadrant is transposed in the S=2.828 example (the red box in the diagram). Maybe it's obvious if one understands it better?

It requires a chunk of linear algebra to understand, but the Wikipedia page has a slightly more detailed explanation: https://en.wikipedia.org/wiki/Bell%27s_theorem#Theorem It's related to the fact that the expected value of A_1 tensor B_1 is negative 1/sqrt(2), whilst the expected value of all other tensor products are positive 1/sqrt(2).

The explanation and table in the Simple English page for this helped me grasp it better. (Although the diagram using green dots only confuses :) )

Greene's book is a fantastic read too!

https://simple.wikipedia.org/wiki/Bell%27s_theorem

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#14

Post by hnax »

GistNoesis wrote: post #9

The Bell's Inequalities are a test for the capacity for inductive reasoning of the pupil. If the pupil succeed he is not to be admitted to join the ranks of quantum physicist. You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore y…

This 'crackpot physicist' is still alive and kicking and, indeed, as per the (analytical) induction requirement to make the case, his work deserves a careful reading to assume the geometric algebraic understanding of QM (for the 'crackpot's latest, see: https://www.linkedin.com/posts/joy-christian-oxford_comment-...)

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#15

Post by GistNoesis »

eigenket wrote: post #12
GistNoesis wrote: post #9

The Bell's Inequalities are a test for the capacity for inductive reasoning of the pupil. If the pupil succeed he is not to be admitted to join the ranks of quantum physicist. You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore y…

> You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore you conclude that the world is non-local. If you do this you're doing a bad job at being a teacher. The way the argument should go is you start with a list of assumptions (of…

>If you do this you're doing a bad job at being a teacher.

If this wasn't sufficiently clear, I am not a teacher, I am a crackpot physicist.

Hint : Listing the assumptions doesn't work. This is what I call the three-card monte argument. The ball is not under one of the three goblets, the ball is in the sleeve of the magician.

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#16

Post by eigenket »

GistNoesis wrote: post #15
eigenket wrote: post #12

Earlier quoted context omitted.

> You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore you conclude that the world is non-local. If you do this you're doing a bad job at being a teacher. The way the argument should go is you start with a list of assumptions (of…

>If you do this you're doing a bad job at being a teacher. If this wasn't sufficiently clear, I am not a teacher, I am a crackpot physicist. Hint : Listing the assumptions doesn't work. This is what I call the three-card monte argument. The ball is not under one of the three goblets, the ball is in the sleeve of the magician.

> Hint : Listing the assumptions doesn't work

Why not?

I could see that it might not if you are not clear about your assumptions

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#17

Post by GistNoesis »

gus_massa wrote: post #10
GistNoesis wrote: post #9

The Bell's Inequalities are a test for the capacity for inductive reasoning of the pupil. If the pupil succeed he is not to be admitted to join the ranks of quantum physicist. You usually show a pupil the problem with classical probabilities, and show that you can't violate Bell's Inequalities, then you show that Quantum Mechanics managed to replicated the observed probabilities using a non-local way, and therefore y…

> and therefore you conclude that the world is non-local. No, Bell's inequality has a few sensible assumtions, like locality. The conclusion is that at least one of them is wrong and real world is a sensible one :(. By the way, there is this crazy thing call QM that nobody likes but gives accurate results.

Just because there is a way, doesn't make it the only way.

>but gives accurate results.

Giving accurate results is missing the point.

Hint: The point is understanding how nature's does it.

Here is the Chesterton's fence implied by Bell's Inequality :

Lemma: There exist a local classical simulator that allows to simulate a universe that behaves according to the probabilities of QM.

Corollary : we can simulate "fast" a universe which behaves (in law) exactly like our universe.

Nota Bene : This doesn't mean we can compute QM probabilities fast, (we can't), although one way of computing them would be to use Montecarlo estimation on various instances of universe simulations.

The question is not whether to lift the fence, or how to lift the fence, the question is how are numerical biological instabilities handled.

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#18

Post by gus_massa »

jahnu wrote: post #13
n4r9 wrote: post #7

Earlier quoted context omitted.

It requires a chunk of linear algebra to understand, but the Wikipedia page has a slightly more detailed explanation: https://en.wikipedia.org/wiki/Bell%27s_theorem#Theorem It's related to the fact that the expected value of A_1 tensor B_1 is negative 1/sqrt(2), whilst the expected value of all other tensor products are positive 1/sqrt(2).

The explanation and table in the Simple English page for this helped me grasp it better. (Although the diagram using green dots only confuses :) ) Greene's book is a fantastic read too! https://simple.wikipedia.org/wiki/Bell%27s_theorem

> Although the diagram using green dots only confuses :)

It's very confusing. In particular it does not say that the box have 3 doors until the middle of the explanations. Also, I don't find the example very similar to the Bell's Inequality.

Moreover, I expect in a quantum system that when both open the same door they get the same result (or the oposite) so in a quantum system I expect that when both open the same door they get 100% (or 0%) agreement, so insted of 50% I expect 1/3 * 100% + 2/3 * 50 % = 66% (or 1/3 * 0% + 2/3 * 50 % = 33%).

Anyway, in some versions of the Bell's Inequality the doors of the boxes are "misalignment" so on box has white-gray-black doors and the other has another ser of colors. let's say creme-pink-brown doors. You never have a 100% or 0% of coincidences of the results.

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#19

Post by GistNoesis »

eigenket wrote: post #16
GistNoesis wrote: post #15

Earlier quoted context omitted.

>If you do this you're doing a bad job at being a teacher. If this wasn't sufficiently clear, I am not a teacher, I am a crackpot physicist. Hint : Listing the assumptions doesn't work. This is what I call the three-card monte argument. The ball is not under one of the three goblets, the ball is in the sleeve of the magician.

> Hint : Listing the assumptions doesn't work Why not? I could see that it might not if you are not clear about your assumptions

It's circular reasoning, hidden in the definition of your assumptions. By defining not clearly what a measurement is and observations are.

You must let the cat step out of the box your definitions put you in.

You have infinite freedom in your choices of definitions, listing assumptions is creating a false dichotomy. Especially when doing so conclude to exclude the most probable assumption : Locality.

Preserve locality, and find another self consistent theory which define properly what according to it a measurement is, an not take measurement and observations as axioms.

Re: Breaking Bell's Inequality with Monte Carlo Simulations in Python

#20

Post by 0cf8612b2e1e »

eigenket wrote: post #11
n4r9 wrote: post #6

This is closely related to my PhD. It was many years ago but if I remember rightly there is no need for the assumption of determinism - Bell Inequalities hold just as well for random local hidden variables. Simulating the correlations with computer programs is an interesting idea, partly because it challenges to those who still believe in a "local" reality to demonstrate Bell Inequality violations in distributed clas…

You know something has gone horrifically badly when a paper begins with > This reply paper should be read as a continuation of my previous reply paper [1], which is a reply published in this journal to a previous critique of one of my papers We're way too deep in replies now, and anyone who values their time should get out now.

On the other hand, academics slap fights are magnificently petty to behold.

  In any dispute the intensity of feeling is inversely proportional to the value of the issues at stake. That is why academic politics are so bitter.
Post reply on HN