SciPhysicsForums.com

This is a challenge to anyone to write a computer simulation of a Bell-CHSH type experiment which reliably violates the Bell-CHSH inequality and which does not violate locality.

The program runs on a single computer. The rules are designed so that no one has to check the code. One just has to check that it satisfies the specifications listed below.

The program must accept as input two streams of settings “1” or “2” of equal length
It will output two streams of outcomes “+1” or “-1” of the same length
The length can be any positive integer
The program must have a “set seed” facility

It must have the following properties:

When given the same seed and the same input streams it produces the same outcomes
When only Alice’s last input is changed then Bob’s last output does not change
When only Bob’s last input is changed then Alice’s last output does not change
When two inputs are added two new outputs are generated but the earlier outputs do not change

The programmer should aim to get a large positive correlation on one setting combination but otherwise a large negative correlation

The programmer wins my challenge with one correlation larger than 0.8 and three correlations smaller than -0.8

This must be attained in runs of length 10 000 with settings supplied by me (simulated fair coin tosses)

It must be reproducible (several testers using different computers and freely choosing different seeds must confirm)

Comments and questions are welcome!

Yeah, and I will give 64 dollars to anyone that can pass the above challenge using quantum theory.
.

FrediFizzx wrote:Yeah, and I will give 64 dollars to anyone that can pass the above challenge using quantum theory.
.

Very important clarification. The programmer can use any theory they like, including Quantum Theory to write their simulation. In fact the use of quantum computers is encouraged.

minkwe wrote:

FrediFizzx wrote:Yeah, and I will give 64 dollars to anyone that can pass the above challenge using quantum theory.
.

Very important clarification. The programmer can use any theory they like, including Quantum Theory to write their simulation. In fact the use of quantum computers is encouraged.

Quantum computers may be used. Notice the rules about reproducibility.

The programmer can use any theory they like. The delivered computer program is a black box. Nobody has to look at the code.

Thanks for your comments! I think you agree that this challenge is one I can safely bet 64 thousand Euro on.

Another point is of course that the companion challenge to quantum technology engineers has not yet been met. We have not yet had a Bell-type experiment which was free of experimental loopholes, had a decent run length, and got S clearly and cleanly above 2.4

Oops, I saw a typo.

The programmer wins my challenge with one correlation larger than 0.6 and three correlations smaller than -0.6

In the O.P. I wrote 0.8, thinking of half of the correlation_plus_one = probability of equal outcomes.

gill1109 wrote:Oops, I saw a typo.

The programmer wins my challenge with one correlation larger than 0.6 and three correlations smaller than -0.6

In the O.P. I wrote 0.8, thinking of half of the correlation_plus_one = probability of equal outcomes.

No one cares about your silly challenge because you are stuck in a quandary with it.
.

FrediFizzx wrote:

gill1109 wrote:Oops, I saw a typo.
The programmer wins my challenge with one correlation larger than 0.6 and three correlations smaller than -0.6
In the O.P. I wrote 0.8, thinking of half of the correlation_plus_one = probability of equal outcomes.

No one cares about your silly challenge because you are stuck in a quandary with it.

“Quandrary”, Fred. I am not stuck in one one. And I clearly live in a different bubble from you.

You say what you say because you desperately want to win my challenge and you think that if you try hard enough you will work out how to do it.

Good luck! Keep on trying!

“Minkwe”: do you agree that it is impossible to win my challenge?

We should formulate a companion challenge to the quantum experimentalists. Then we could start taking bets as to whether or not they will do it, in, say, ten years. I think it’s 50-50. So I’m willing to bet a fistful of dollars on that. Either way.

gill1109 wrote:

FrediFizzx wrote:

gill1109 wrote:Oops, I saw a typo.
The programmer wins my challenge with one correlation larger than 0.6 and three correlations smaller than -0.6
In the O.P. I wrote 0.8, thinking of half of the correlation_plus_one = probability of equal outcomes.

No one cares about your silly challenge because you are stuck in a quandary with it.

“Quandrary”, Fred. I am not stuck in one one. And I clearly live in a different bubble from you. ...

Yes, you are stuck in a quandary. You are saying that NOTHING can beat the challenge except that the experiments and Nature have already shot you down so obviously wrong. You are pretty foolish to be risking that much money on something that is so obviously wrong. I'm not really interested anymore in it but someone some day is going to put you in the poor house.
.

.
I have a counter challenge and I am happy to give a promissory note of 64 million dollars, right here on this very post, to anyone who meets my challenge, which I have set out below:

Forget quantum mechanics or Bell. Classical mechanics is known to us and has been perfected for the past 300 years. For classical systems, such as weather systems, it is a perfectly valid mechanics. Given the correct initial conditions, it is a deterministic, dispersion-free mechanics. If you accept my challenge, then I will give you the initial conditions, which are the exact meteorological conditions at the time and date of my posting this message. My challenge is very simple. Provide me a Yes or No answer to the following question: Will it be raining precisely at 3:00 PM, GMT, on the Christmas day, near the main gate of the Clarendon Laboratory, located on Parks Road within the Science Area in Oxford, England, United Kingdom? Yes, or No.

If you give me the correct answer, which I will verify on the Christmas day, then you will earn the above promissory note from me. But if you get your answer wrong, then you owe me 64 million dollars! Note that my offer of 64 million dollars ends at noon, GMT, on the 15th of December 2020. Make your prediction before that date and have your 64 million dollars ready.
.

FrediFizzx wrote:Yes, you are stuck in a quandary. You are saying that NOTHING can beat the challenge except that the experiments and Nature have already shot you down so obviously wrong. You are pretty foolish to be risking that much money on something that is so obviously wrong. I'm not really interested anymore in it but someone some day is going to put you in the poor house.

Fred, the experiments and Nature don't satisfy the conditions which I listed.

I suggest you read them. I asked for reproducibility. I would be delighted if you could win the challenge. You would thereby win the Nobel prize.

Joy Christian wrote:.
I have a counter challenge and I am happy to give a promissory note of 64 million dollars, right here on this very post, to anyone who meets my challenge, which I have set out below:

Forget quantum mechanics or Bell. Classical mechanics is known to us and has been perfected for the past 300 years. For classical systems, such as weather systems, it is a perfectly valid mechanics. Given the correct initial conditions, it is a deterministic, dispersion-free mechanics. If you accept my challenge, then I will give you the initial conditions, which are the exact meteorological conditions at the time and date of my posting this message. My challenge is very simple. Provide me a Yes or No answer to the following question: Will it be raining precisely at 3:00 PM, GMT, on the Christmas day, near the main gate of the Clarendon Laboratory, located on Parks Road within the Science Area in Oxford, England, United Kingdom? Yes, or No.

If you give me the correct answer, which I will verify on the Christmas day, then you will earn the above promissory note from me. But if you get your answer wrong, then you owe me 64 million dollars! Note that my offer of 64 million dollars ends at noon, GMT, on the 15th of December 2020. Make your prediction before that date and have your 64 million dollars ready.
.

Joy, you are unable to give us the initial conditions, because it is the classical physics state of every particle in the entire Earth's atmosphere, and more. Moreover, if you give it at a sufficiently fine-grained level, it will involve quantum interactions between different particles. Moreover, the oceans and the sun and a whole lot more also come in. Even if you gave us all those conditions, we would basically have to build a new solar system in order to have an analogue computer which could actually do the computation, because we certainly would not be able to do it with existing computers. Finally, you are unable to pay 64 million dollars, and I am unable to do it, too. So I can make the prediction right now, but the promissory notes, either way, will have absolutely no value whatever.

By the way, I have twice made a financial contribution to your Oxford institute, but I never received a receipt.

.
The point of my counter-challenge was to demonstrate that Gill's challenge is as worthless and ridiculous as my weather challenge is. Our mechanics, classical or quantum, allow us to make only in-principle predictions of specific definite outcomes even in as simple a case as the weather. The answer can be "Yes" or "No", but we cannot say exactly which one in a given specific scenario. This makes Gill's so-called challenge a silly joke that he himself does not get.
.

gill1109 wrote:

FrediFizzx wrote:Yes, you are stuck in a quandary. You are saying that NOTHING can beat the challenge except that the experiments and Nature have already shot you down so obviously wrong. You are pretty foolish to be risking that much money on something that is so obviously wrong. I'm not really interested anymore in it but someone some day is going to put you in the poor house.

Fred, the experiments and Nature don't satisfy the conditions which I listed.

I suggest you read them. I asked for reproducibility. ...

Another reason your challenge is silly. Reproducibility can easily be dispensed with. The original goal was to simulate a typical EPR experiment. You've tacked on crap that doesn't matter. Very silly. You are still stuck in a quandary.
.

Let's suppose for a minute that it is possible for quantum mechanics to predict the individual event by event outcomes for A and B. How might one go about doing that? Well, we would have to construct measurement functions for A and B,




Well, we know from Bell those functions give the straight line triangle correlation. But that would be the QM prediction for an event by event simulation of an EPR scenario which of course matches other local theory predictions. Of course, the experiments say that it is probably wrong.
.

FrediFizzx wrote:

gill1109 wrote:

FrediFizzx wrote:Yes, you are stuck in a quandary. You are saying that NOTHING can beat the challenge except that the experiments and Nature have already shot you down so obviously wrong. You are pretty foolish to be risking that much money on something that is so obviously wrong. I'm not really interested anymore in it but someone some day is going to put you in the poor house.

Fred, the experiments and Nature don't satisfy the conditions which I listed.
I suggest you read them. I asked for reproducibility. ...

Another reason your challenge is silly. Reproducibility can easily be dispensed with. The original goal was to simulate a typical EPR experiment. You've tacked on crap that doesn't matter. Very silly. You are still stuck in a quandary.

No Fred, my original goal is to interact with people who think that Bell’s theorem is wrong, and who believe they have a counterexample. I challenge them to convert their theory into a computer package. I don’t want to spend hours studying the code. I want to be able to run a few tests in order to verify whether they have simulated a properly conducted Bell-CHSH type experiment, including implementation of the separation between measurement stations.

*Your goal* is to simulate an EPR-B experiment, I believe.

Do you want to simulate one with “separated measurements’, yes or no?

Do you think you can do it *and* reproduce the negative cosine?

Do you think you can do it with settings supplied externally, one binary outcome at each measurement station for each measurement setting?

If so, you can win my challenge.

Joy Christian wrote:.
The point of my counter-challenge was to demonstrate that Gill's challenge is as worthless and ridiculous as my weather challenge is. Our mechanics, classical or quantum, allow us to make only in-principle predictions of specific definite outcomes even in as simple a case as the weather. The answer can be "Yes" or "No", but we cannot say exactly which one in a given specific scenario. This makes Gill's so-called challenge a silly joke that he himself does not get.
.

So why do your recent publications, Joy, include computer programs and Monte Carlo simulations? I think you call them “event based” and say they do not use post-selection or other tricks. You say they demonstrate that Bell’s theorem is false.

I guess they are just a scientific hoax then, i.e. a joke. A good joke, not a silly joke! I respond with my own joke.

Thanks everyone for all the responses. I will write up the challenge in a publication and submit it somewhere.

PS to Fred:

FrediFizzx wrote:Let's suppose for a minute that it is possible for quantum mechanics to predict the individual event by event outcomes for A and B. How might one go about doing that? Well, we would have to construct measurement functions for A and B,


Well, we know from Bell those functions give the straight line triangle correlation. But that would be the QM prediction for an event by event simulation of an EPR scenario which of course matches other local theory predictions. Of course, the experiments say that it is probably wrong.

Please explain what you mean by your limit notation. How can s converge to a limit which depends on s ?

gill1109 wrote:PS to Fred:

FrediFizzx wrote:Let's suppose for a minute that it is possible for quantum mechanics to predict the individual event by event outcomes for A and B. How might one go about doing that? Well, we would have to construct measurement functions for A and B,


Well, we know from Bell those functions give the straight line triangle correlation. But that would be the QM prediction for an event by event simulation of an EPR scenario which of course matches other local theory predictions. Of course, the experiments say that it is probably wrong.

Please explain what you mean by your limit notation. How can s converge to a limit which depends on s ?

It is all exactly explained by the most RHS expressions. It is some pretty simple math.
.

gill1109 wrote: ...
*Your goal* is to simulate an EPR-B experiment, I believe.
...

No. My goal is to demonstrate to everyone how silly your junk theory is.
.

gill1109 wrote: ...
No Fred, my original goal is to interact with people who think that Bell’s theorem is wrong, and who believe they have a counterexample. I challenge them to convert their theory into a computer package. I don’t want to spend hours studying the code. I want to be able to run a few tests in order to verify whether they have simulated a properly conducted Bell-CHSH type experiment, including implementation of the separation between measurement stations. ...

Sorry, your silly challenge is REJECTED. You are not actually doing "a properly conducted Bell-CHSH type experiment". You've tried to escape the quandary you are stuck in but strip away the nonsense and you are still stuck in that quandary.
.

FrediFizzx wrote:

gill1109 wrote:PS to Fred:

FrediFizzx wrote:Let's suppose for a minute that it is possible for quantum mechanics to predict the individual event by event outcomes for A and B. How might one go about doing that? Well, we would have to construct measurement functions for A and B,


Well, we know from Bell those functions give the straight line triangle correlation. But that would be the QM prediction for an event by event simulation of an EPR scenario which of course matches other local theory predictions. Of course, the experiments say that it is probably wrong.

Please explain what you mean by your limit notation. How can s converge to a limit which depends on s ?

It is all exactly explained by the most RHS expressions. It is some pretty simple math.
.

The RHS only tells us what you want the result to be of evaluating that limit. But the limit expression itself is illegal. Go back to school, Fred!

FrediFizzx wrote:

gill1109 wrote: ...
*Your goal* is to simulate an EPR-B experiment, I believe.
...

No. My goal is to demonstrate to everyone how silly your junk theory is.
.

You are failing miserably.

gill1109 wrote:

FrediFizzx wrote:

gill1109 wrote: ...
*Your goal* is to simulate an EPR-B experiment, I believe.
...

No. My goal is to demonstrate to everyone how silly your junk theory is.
.

You are failing miserably. ...

I'm not the one stuck in a quandary.
.

gill1109 wrote:

FrediFizzx wrote:

gill1109 wrote:PS to Fred:

FrediFizzx wrote:Let's suppose for a minute that it is possible for quantum mechanics to predict the individual event by event outcomes for A and B. How might one go about doing that? Well, we would have to construct measurement functions for A and B,


Well, we know from Bell those functions give the straight line triangle correlation. But that would be the QM prediction for an event by event simulation of an EPR scenario which of course matches other local theory predictions. Of course, the experiments say that it is probably wrong.

Please explain what you mean by your limit notation. How can s converge to a limit which depends on s ?

It is all exactly explained by the most RHS expressions. It is some pretty simple math.
.

The RHS only tells us what you want the result to be of evaluating that limit. But the limit expression itself is illegal. Go back to school, Fred!

It is quite amazing that a mathematician doesn't understand how a limit substitution process works! Ah, but 'tis a typical Gill distraction from the main result that QM might predict the straight line triangle correlation for event by event outcomes.
.