A Completelly Local and Realistic Simulation

Foundations of physics and/or philosophy of physics, and in particular, posts on unresolved or controversial issues

Re: A Completelly Local and Realistic Simulation

Postby jreed » Tue Feb 02, 2021 10:49 am

FrediFizzx wrote:Recently, there has been a lot of news about some really aweful and dangerous conspiracy theories. Well, I have a good conspiracy theory pertaining to the so-called Bell test experiments. :D I have further modified the simulation that started this thread. First, I took out the Pearle function as it is not all that necessary. Second, the hidden variable is now a function of the singlet vector e which is perhaps more appropriate. Third, I took out the absolute value function on the A and B functions for the constraints to highlight the fact that there is also a negative HV lambda. Fourth, you will see the conspiracy for the a and b detection vectors. During the constraints, those vectors are replaced by another set of random integers from 1 to 360 degrees. The main point of all this is that you can't tell from the results that anything is out of the ordinary. So, perhaps something similar to this is going on for the test experiments. And the result for 10 million trials at one degree resolution is,

Image

Here is a PDF of the Mathematica program along with the notebook file for those that might be interested,

EPRsims/newCS-1-forum.pdf
EPRsims/newCS-1-forum.nb
.


Fred, I took this program and modified it so I could set lambda to zero. This results in the familiar triangle shape with a CHSH of 1.97. The conclusion is that this program is using the detection loophole to generate the cosine curve you are displaying. This all dates back to minkwe's use of the loophole in EPR Simple.
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Tue Feb 02, 2021 12:17 pm

jreed wrote:Fred, I took this program and modified it so I could set lambda to zero. This results in the familiar triangle shape with a CHSH of 1.97. The conclusion is that this program is using the detection loophole to generate the cosine curve you are displaying. This all dates back to minkwe's use of the loophole in EPR Simple.

John, almost all the events are detected so it is not the detection loophole. I believe they call it the conspiracy loophole; that is why I was mentioning conspiracy theories at the beginning. Yes, setting the HV to zero disables the constraints on A and B so you are just back to Sign[a.s} and Sign{b.s} which of course gives the straight line triangle shape plot.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Tue Feb 02, 2021 2:11 pm

FrediFizzx wrote:
jreed wrote:Fred, I took this program and modified it so I could set lambda to zero. This results in the familiar triangle shape with a CHSH of 1.97. The conclusion is that this program is using the detection loophole to generate the cosine curve you are displaying. This all dates back to minkwe's use of the loophole in EPR Simple.

John, almost all the events are detected so it is not the detection loophole. I believe they call it the conspiracy loophole; that is why I was mentioning conspiracy theories at the beginning. Yes, setting the HV to zero disables the constraints on A and B so you are just back to Sign[a.s} and Sign{b.s} which of course gives the straight line triangle shape plot.
.

The main point of this exercise is to demonstrate that perhaps somehow Nature is tricking the experimenters.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby gill1109 » Wed Feb 03, 2021 2:34 am

FrediFizzx wrote:
FrediFizzx wrote:
jreed wrote:Fred, I took this program and modified it so I could set lambda to zero. This results in the familiar triangle shape with a CHSH of 1.97. The conclusion is that this program is using the detection loophole to generate the cosine curve you are displaying. This all dates back to minkwe's use of the loophole in EPR Simple.

John, almost all the events are detected so it is not the detection loophole. I believe they call it the conspiracy loophole; that is why I was mentioning conspiracy theories at the beginning. Yes, setting the HV to zero disables the constraints on A and B so you are just back to Sign[a.s} and Sign{b.s} which of course gives the straight line triangle shape plot.
.

The main point of this exercise is to demonstrate that perhaps somehow Nature is tricking the experimenters.
.

Neat work, Fred!

Indeed, Nature could be tricking the experimenters in many experiments. Some experimenters don’t really know what they are doing: they just get excited when they’ve managed to find a negative cosine.
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Thu Feb 04, 2021 8:53 am

gill1109 wrote:
FrediFizzx wrote:
FrediFizzx wrote:John, almost all the events are detected so it is not the detection loophole. I believe they call it the conspiracy loophole; that is why I was mentioning conspiracy theories at the beginning. Yes, setting the HV to zero disables the constraints on A and B so you are just back to Sign[a.s} and Sign{b.s} which of course gives the straight line triangle shape plot.
.

The main point of this exercise is to demonstrate that perhaps somehow Nature is tricking the experimenters.
.

Neat work, Fred!

Indeed, Nature could be tricking the experimenters in many experiments. Some experimenters don’t really know what they are doing: they just get excited when they’ve managed to find a negative cosine.

Thanks. Of course this extends from the other thread where I was doing CHSH. I was really surprised that having another set of angles during the constraints worked. Still not sure exactly why. However, no one noticed the crazy function for the hidden variable. If plotted out it looks like this,

Image

A bunch of oscillations from 0 to 0.5 when the singlet vector, e, goes from 0 to 360 degrees.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby jreed » Thu Feb 04, 2021 11:12 am

Fred, you are computing the cosine in the lambda function using angles in degrees. You should use angles in radians since all the trig functions in Mathematica are set up for radians. I have corrected this in my version, and it doesn't seem to matter. You must have spent a lot of time coming up with that cosine output curve at the end. I don't understand what this calculation represents physically, but that cosine curve is really nice.
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Thu Feb 04, 2021 1:49 pm

jreed wrote:Fred, you are computing the cosine in the lambda function using angles in degrees. You should use angles in radians since all the trig functions in Mathematica are set up for radians. I have corrected this in my version, and it doesn't seem to matter. You must have spent a lot of time coming up with that cosine output curve at the end. I don't understand what this calculation represents physically, but that cosine curve is really nice.

John, actually I am computing the cosine in the HV function using radians but feeding angles in degrees to it as if they are radians. That is on purpose as it helps to smooth out the very tail ends of the negative cosine curve. Not sure why yet. The calculation is a thought experiment on possibly how experimenters might be being tricked by Nature. I haven't seen a lot of simulations that use the conspiracy loophole concept.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Sat Feb 06, 2021 9:10 am

I would like to stress here that this conspiracy loophole is NOT a loophole to Bell's theory since Bell's theory is shot down. This conspiracy loophole is a loophole to Gill's theory.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby jreed » Sat Feb 06, 2021 10:38 am

FrediFizzx wrote:
jreed wrote:Fred, you are computing the cosine in the lambda function using angles in degrees. You should use angles in radians since all the trig functions in Mathematica are set up for radians. I have corrected this in my version, and it doesn't seem to matter. You must have spent a lot of time coming up with that cosine output curve at the end. I don't understand what this calculation represents physically, but that cosine curve is really nice.

John, actually I am computing the cosine in the HV function using radians but feeding angles in degrees to it as if they are radians. That is on purpose as it helps to smooth out the very tail ends of the negative cosine curve. Not sure why yet. The calculation is a thought experiment on possibly how experimenters might be being tricked by Nature. I haven't seen a lot of simulations that use the conspiracy loophole concept.
.

The conspiracy loophole, as I understand it, is that Bob knows what angle Alice has her detector set at. He can then adjust his angle to give the cosine output. What represents this concept in your simulation?
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Sat Feb 06, 2021 1:22 pm

jreed wrote:
FrediFizzx wrote:
jreed wrote:Fred, you are computing the cosine in the lambda function using angles in degrees. You should use angles in radians since all the trig functions in Mathematica are set up for radians. I have corrected this in my version, and it doesn't seem to matter. You must have spent a lot of time coming up with that cosine output curve at the end. I don't understand what this calculation represents physically, but that cosine curve is really nice.

John, actually I am computing the cosine in the HV function using radians but feeding angles in degrees to it as if they are radians. That is on purpose as it helps to smooth out the very tail ends of the negative cosine curve. Not sure why yet. The calculation is a thought experiment on possibly how experimenters might be being tricked by Nature. I haven't seen a lot of simulations that use the conspiracy loophole concept.
.

The conspiracy loophole, as I understand it, is that Bob knows what angle Alice has her detector set at. He can then adjust his angle to give the cosine output. What represents this concept in your simulation?

That is a non-local loophole that doesn't apply to local HV theories. My conspiracy loophole is completely local and 100 percent realistic. In my model, a is replaced by ar and b is replaced by br during the constraints. That is the conspiracy.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Sun Feb 07, 2021 7:31 pm

Ok, I've fixed the screwy way I was doing the HV function. Now, I'm taking in the singlet vector angles as degrees but multiplying the result by 180/pi so we have,



So, the hidden variable still oscilates about 57 times between 0 and 1/2 during the 0 to 360 for the singlet vector angle.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby gill1109 » Sun Feb 07, 2021 11:16 pm

jreed wrote:
FrediFizzx wrote:
jreed wrote:Fred, you are computing the cosine in the lambda function using angles in degrees. You should use angles in radians since all the trig functions in Mathematica are set up for radians. I have corrected this in my version, and it doesn't seem to matter. You must have spent a lot of time coming up with that cosine output curve at the end. I don't understand what this calculation represents physically, but that cosine curve is really nice.

John, actually I am computing the cosine in the HV function using radians but feeding angles in degrees to it as if they are radians. That is on purpose as it helps to smooth out the very tail ends of the negative cosine curve. Not sure why yet. The calculation is a thought experiment on possibly how experimenters might be being tricked by Nature. I haven't seen a lot of simulations that use the conspiracy loophole concept.
.

The conspiracy loophole, as I understand it, is that Bob knows what angle Alice has her detector set at. He can then adjust his angle to give the cosine output. What represents this concept in your simulation?

Bell’s theorem says that quantum mechanics is incompatible with locality+realism+freedom. Freedom, aka no-conspiracy, is the freedom to pick settings at will independently of the physics going on in source+transmission lines+detectors. As I understand it, in Fred’s model, the freely chosen settings are overwritten with settings created by the hidden variables in the system.

No conspiracy is also an assumption of reductionism. One can separate the universe into pieces, study them separately, then put the pieces together and understand the whole. The converse is some kind of holism. Everything depends on everything else, you cannot understand the whole by reducing it to separate components. David Bohm took that route. It fits many people’s spiritual yearnings but it is hard to see how it can be a basis for sound physics. A similar idea is that we are just living in a simulation, nothing is real at all, and our idea that we have agency is an illusion. The world is how it appears to be because whoever wrote the simulation program wanted us to experience that. Take the red pill!

Indeed it is possible that many earlier published experimental results are actually caused by mechanisms like that in Fred’s model. That’s why physicists adopted Bell’s ideas for how to perform a loophole-free experiment.

The results can always be explained away by claiming conspiracy, but the idea of a loophole free experiment is to make that explanation as far-fetched as possible. Do you really want to believe that Alice’s coin toss outcome which she uses to choose in which of two positions to set a toggle switch was physically predetermined to perfectly match hidden variables located in a photo-detector 50 miles away? And so as to exactly reproduce quantum correlations but not to exceed them? I’m sure Fred can easily tune his parameters to violate Tsirelson’s bound. Thereby experimentally disproving quantum mechanics!
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: A Completelly Local and Realistic Simulation

Postby jreed » Mon Feb 08, 2021 8:26 am

FrediFizzx wrote:Ok, I've fixed the screwy way I was doing the HV function. Now, I'm taking in the singlet vector angles as degrees but multiplying the result by 180/pi so we have,



So, the hidden variable still oscilates about 57 times between 0 and 1/2 during the 0 to 360 for the singlet vector angle.
.

I have found that just setting lambda=0.25 for all i gives an equally good cosine output. 0.25 is approximately the mean of the lambda function you are using. It is used as a switch that determines which detector angle is used (in Alice's case, a or ar). The value of |Cos(a - s)| is compared to lambda. If this is less than lambda, a is used to compute the value of Alice's detector, A. If greater than lambda, a new value, ar, is used to compute A. This new value ar is different than the value a, which was compared to lambda. ar is found as a new random angle variable, so there must be a second detector somewhere. All these values are included in the statistics that follow.
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Mon Feb 08, 2021 8:57 am

jreed wrote:
FrediFizzx wrote:Ok, I've fixed the screwy way I was doing the HV function. Now, I'm taking in the singlet vector angles as degrees but multiplying the result by 180/pi so we have,



So, the hidden variable still oscilates about 57 times between 0 and 1/2 during the 0 to 360 for the singlet vector angle.
.

I have found that just setting lambda=0.25 for all i gives an equally good cosine output. 0.25 is approximately the mean of the lambda function you are using. It is used as a switch that determines which detector angle is used (in Alice's case, a or ar). The value of |Cos(a - s)| is compared to lambda. If this is less than lambda, a is used to compute the value of Alice's detector, A. If greater than lambda, a new value, ar, is used to compute A. This new value ar is different than the value a, which was compared to lambda. ar is found as a new random angle variable, so there must be a second detector somewhere. All these values are included in the statistics that follow.

Setting lambda to 0.25 messes up the peaks somewhat.

Image

Plus then you have to explain where that 0.25 comes from physically. With my method, you only need to know the single vector e, ar, and br in addition to already knowing a and b to predict the A and B outcomes.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby jreed » Mon Feb 08, 2021 11:01 am

Well, you have to explain where that function lambda comes from physically, which you also need for your method. Just by running a few trials, I found that lambda = 0.22 works better. After all this is just a Mathematica game, isn't it?
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

Re: A Completelly Local and Realistic Simulation

Postby gill1109 » Mon Feb 08, 2021 11:12 am

jreed wrote:Well, you have to explain where that function lambda comes from physically, which you also need for your method. Just by running a few trials, I found that lambda = 0.22 works better. After all this is just a Mathematica game, isn't it?

I recently saw a nice paper which investigated just how much non-locality, or just how much conspiracy, is needed to fake quantum non-locality, ie a Bell inequality violation of a certain size. It got some rather nice results. I think those results are related to what you are finding here. I will look it up.
gill1109
Mathematical Statistician
 
Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Mon Feb 08, 2021 2:49 pm

jreed wrote:Well, you have to explain where that function lambda comes from physically, which you also need for your method. Just by running a few trials, I found that lambda = 0.22 works better. After all this is just a Mathematica game, isn't it?

What also works well is



Image

Sure, it is just a game. Play away! Maybe you will hit the jackpot. :D
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby jreed » Mon Feb 08, 2021 5:18 pm

gill1109 wrote:
jreed wrote:Well, you have to explain where that function lambda comes from physically, which you also need for your method. Just by running a few trials, I found that lambda = 0.22 works better. After all this is just a Mathematica game, isn't it?

I recently saw a nice paper which investigated just how much non-locality, or just how much conspiracy, is needed to fake quantum non-locality, ie a Bell inequality violation of a certain size. It got some rather nice results. I think those results are related to what you are finding here. I will look it up.


I think it has something to do with the detection loophole. The lambda function that Fred is using originally came from Fodje's epr-simple python program. He also had a second function that worked equally well. No explanation was given for these functions, but it was obvious that they threw out some values. In Fred's version, this function is used to call two different random angle variables (a and ar) to compute A. Somehow mixing together these two distributions seems to be equivalent to the detection loophole. Perhaps you can give some insight into how this might work.
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

Re: A Completelly Local and Realistic Simulation

Postby FrediFizzx » Mon Feb 08, 2021 7:48 pm

jreed wrote:
gill1109 wrote:
jreed wrote:Well, you have to explain where that function lambda comes from physically, which you also need for your method. Just by running a few trials, I found that lambda = 0.22 works better. After all this is just a Mathematica game, isn't it?

I recently saw a nice paper which investigated just how much non-locality, or just how much conspiracy, is needed to fake quantum non-locality, ie a Bell inequality violation of a certain size. It got some rather nice results. I think those results are related to what you are finding here. I will look it up.


I think it has something to do with the detection loophole. The lambda function that Fred is using originally came from Fodje's epr-simple python program. He also had a second function that worked equally well. No explanation was given for these functions, but it was obvious that they threw out some values. In Fred's version, this function is used to call two different random angle variables (a and ar) to compute A. Somehow mixing together these two distributions seems to be equivalent to the detection loophole. Perhaps you can give some insight into how this might work.

Nope. Not really equivalent to the detection loophole. Out of 3 million events 2999983 are detected. Of course this model was derived from the detection loophole model but instead of having A = B = 0 during the constraints, I originally set a = b = 0 during the constraints. However, all the events during the constraints were piling up at theta = 0 and cancelling each other out. But with this new model, all the events are spread out like they should be and indeed it remains a mystery as to why the mixing works. I'm still analyzing that.
.
FrediFizzx
Independent Physics Researcher
 
Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: A Completelly Local and Realistic Simulation

Postby jreed » Tue Feb 09, 2021 8:10 am

I agree with you. In the version I'm using, all events are detected. I did 1,000,000 trials, and 835,710 went into the a detector, 164,290 into the ar detector. None are missing. It's not the detection loophole. I get similar results with lambda set at a constant 0.22. You may have discovered a new loophole, or maybe a disproof of Bell's theorem. It's very interesting.
jreed
 
Posts: 176
Joined: Mon Feb 17, 2014 5:10 pm

PreviousNext

Return to Sci.Physics.Foundations

Who is online

Users browsing this forum: No registered users and 68 guests

cron
CodeCogs - An Open Source Scientific Library