Google quantum supremacy paper

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

Google quantum supremacy paper

Postby Dirkman » Tue Sep 24, 2019 9:56 pm

So I guess you've already heard about the paper that was posted https://www.inverse.com/article/59507-full-quantum-supremacy-paper and pulled back by them , announcing quantum supremacy (completing a task that wouldve required millennia for a supercomputer)

Why do you think they pulled it back?
Dirkman
 
Posts: 24
Joined: Sat Aug 22, 2015 9:39 pm

Re: Google quantum supremacy paper

Postby Joy Christian » Tue Sep 24, 2019 10:02 pm

Dirkman wrote:So I guess you've already heard about the paper that was posted https://www.inverse.com/article/59507-full-quantum-supremacy-paper and pulled back by them , announcing quantum supremacy (completing a task that wouldve required millennia for a supercomputer)

Why do you think they pulled it back?

Yes, we have heard about the Google paper --- now you see it, now you don't.

Although Google claims to have attained quantum supremacy experimentally, as yet there is no definitive theoretical proof provided for it, either by them or by anyone else.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Heinera » Tue Sep 24, 2019 11:42 pm

Dirkman wrote:Why do you think they pulled it back?


Because those who put it out there very soon were told that the paper is embargoed by the journal Nature until publishing, which is scheduled for October.
Heinera
 
Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Google quantum supremacy paper

Postby Jarek » Thu Sep 26, 2019 12:21 am

Quantum computers are a big problem for hypothetical local realistic models - which would allow for their fast simulation in classical computers ... e.g. for Shor's algorithm breaking asymmetric cryptography we use everywhere.
Advocating local realism, you need to point the obstacle for such simulation ...

Looking at Shor's algorithm ( https://physics.stackexchange.com/quest ... its-crippl ), it literally splits calculation into two branches, enters input in one branch, reads corresponding output in the second branch. Wanting local realistic explanation, the causality needs to go back to the splitting point (classical function) - again requiring accepting time symmetry, which is a big obstacle for classical simulation.

Image
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Joy Christian » Thu Sep 26, 2019 12:29 am

Jarek wrote:
Quantum computers are a big problem for hypothetical local realistic models...

Quantum computers are not a problem for my local-realistic model at all, because my model reproduces ALL quantum correlations exactly as they are predicted by quantum mechanics.

One must never confuse any analitical local-realistic model with computer simulations. If you cannot simulate a local-realistic model, then you are not a good programmer. It is that simple.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Jarek » Thu Sep 26, 2019 12:38 am

So what stops people from simulating Shor's algorithm with your model to break used asymmetric cryptography? It must be fairness and honesty ...
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Joy Christian » Thu Sep 26, 2019 12:49 am

Jarek wrote:
So what stops people from simulating Shor's algorithm with your model to break used asymmetric cryptography?

I am not stopping anyone.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Joy Christian » Thu Sep 26, 2019 1:49 am

***
The supremacy claims by Google are probably false: https://gilkalai.wordpress.com/2019/09/ ... oogle/amp/.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Jarek » Sun Sep 29, 2019 2:19 am

I am not stopping anyone.

Recreating working Shor's algorithm with your model would immediately convince everybody - in contrast to a hundred of theoretical papers.

So maybe just propose how would you realize its building blocks: quantum gates ( https://en.wikipedia.org/wiki/Quantum_logic_gate ) and measurement ?
Claiming to have local realistic model it shouldn't be a problem (?) ... and using it anybody could build such simulation - If looking promising, e.g. I would gladly work on that.

I was optimistic for such local realistic hidden variable models a decade ago, but this optimism was cooled down by Shor: while we have superposition also classically e.g of waves, Shor is able to restrict ensemble of such waves on one branch of calculation, and read result for such restricted ensemble on another branch of calculation - wanting to realize it in local realistic way, the causality needs to go back to the branching moment: this algorithm directly exploits time symmetry.
Can your model overcome this difficulty?
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Joy Christian » Sun Sep 29, 2019 2:53 am

Jarek wrote:
I am not stopping anyone.

Recreating working Shor's algorithm with your model would immediately convince everybody - in contrast to a hundred of theoretical papers.

So maybe just propose how would you realize its building blocks: quantum gates ( https://en.wikipedia.org/wiki/Quantum_logic_gate ) and measurement ?
Claiming to have local realistic model it shouldn't be a problem (?) ... and using it anybody could build such simulation - If looking promising, e.g. I would gladly work on that.

I was optimistic for such local realistic hidden variable models a decade ago, but this optimism was cooled down by Shor: while we have superposition also classically e.g of waves, Shor is able to restrict ensemble of such waves on one branch of calculation, and read result for such restricted ensemble on another branch of calculation - wanting to realize it in local realistic way, the causality needs to go back to the branching moment: this algorithm directly exploits time symmetry.
Can your model overcome this difficulty?

I do not see any difficulty. But then again, I have no interest in any computer algorithms.

But you are welcome to work on the problem you are suggesting. My local-realistic model is openly accessible from the following three published papers:

https://royalsocietypublishing.org/doi/ ... sos.180526

https://ieeexplore.ieee.org/document/8836453

https://link.springer.com/article/10.10 ... 014-2412-2

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Jarek » Sun Sep 29, 2019 3:23 am

Your papers are focused on EPR-like settings. To show that it is more universal, you need to provide constructions of:
- Hadamard gate used to get the initial superpostion,
- gates for calculation - it is sufficient to get CNOT (controlled-NOT), others can be built from it,
- measurement process.
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Joy Christian » Sun Sep 29, 2019 8:29 am

Jarek wrote:Your papers are focused on EPR-like settings. To show that it is more universal, you need to provide constructions of:
- Hadamard gate used to get the initial superposition,
- gates for calculation - it is sufficient to get CNOT (controlled-NOT), others can be built from it,
- measurement process.

I don't have to provide anything. My RSOS paper presents a local-realistic framework for computing ALL quantum correlations without needing a Born rule to predict experimental outcomes.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Jarek » Sun Sep 29, 2019 11:34 pm

Yes, I have seen your "computing ALL quantum correlation" claim a few times (as the only response to all criticism).
If so, it should also agree for setting of Shor's algorithm, factorizing numbers in polynomial time, breaking e.g. RSA - if you want anybody to agree with your belief, this is the ultimate way. QC are the final test for hidden variable models.
You can say that "you don't care", but your actions say just the opposite - your fight with quantum computers clearly shows that you have tried but it didn't work.
It is extremely educative to localize this obstacle: understand where the superiority of QC come from, why we cannot quickly simulate them.
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Joy Christian » Mon Sep 30, 2019 12:02 am

Jarek wrote:Yes, I have seen your "computing ALL quantum correlation" claim a few times (as the only response to all criticism).
If so, it should also agree for setting of Shor's algorithm, factorizing numbers in polynomial time, breaking e.g. RSA - if you want anybody to agree with your belief, this is the ultimate way. QC are the final test for hidden variable models.
You can say that "you don't care", but your actions say just the opposite - your fight with quantum computers clearly shows that you have tried but it didn't work.
It is extremely educative to localize this obstacle: understand where the superiority of QC comes from, why we cannot quickly simulate them.

Because the "superiority [or supremacy] of QC" does not exist. What can be done by quantum computers can be done by classical computers. The hype of QC will evaporate sooner or later.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Jarek » Mon Sep 30, 2019 12:22 am

This is exactly the type of criticism I was talking about - you were not able to get QC with your model, so you neglect QC.
However, they just use quantum formalism, only in a larger scale.
If you are able to realize Hadamard and CNOT gate, also measurement - you could build e.g. Shor's algorithm from them.
Not being able to get Shor means that you cannot realize at least one of these building blocks.

This is extremely educative exercise to understand why you couldn't make working Shor's algorithm with your model (or other hidden variables), while QM formalism allows for that - try localizing the obstacle.
The standard answer is that it is superposition e.g. of 2^n possibilities, but we can get superposition e.g. of waves - the problem is somewhere else.
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Joy Christian » Mon Sep 30, 2019 12:30 am

Jarek wrote:
This is exactly the type of criticism I was talking about - you were not able to get QC with your model, so you neglect QC.

Incorrect. I ignore QC because QC is a hype that will evaporate sooner or later. It is not physics. I am interested only in physics.

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby Jarek » Mon Sep 30, 2019 12:55 am

According to quantum formalism, Shor's algorithm should work - what exactly do you disagree with: QM formalism, or maybe you can point a reason why Shor cannot work in QM formalism?
Jarek
 
Posts: 241
Joined: Tue Dec 08, 2015 1:57 am

Re: Google quantum supremacy paper

Postby Dirkman » Wed Oct 23, 2019 3:50 am

They made it official today.
Dirkman
 
Posts: 24
Joined: Sat Aug 22, 2015 9:39 pm

Re: Google quantum supremacy paper

Postby Joy Christian » Wed Oct 23, 2019 3:55 am

Dirkman wrote:
They made it official today.

That does not change anything.

Ok, it changes something. It consolidates the sociology of false belief in the superiority of quantum computers, if there is such a thing.

PS: The Google paper is already disputed by IBM: https://www.ibm.com/blogs/research/2019 ... supremacy/

IBM: "Because the original meaning of the term “quantum supremacy,” as proposed by John Preskill in 2012, was to describe the point where quantum computers can do things that classical computers can’t, this threshold has not been met."

***
Joy Christian
Research Physicist
 
Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Google quantum supremacy paper

Postby local » Wed Oct 23, 2019 9:19 am

Very interesting! Thank you for the link.

I share your view that "quantum computation" is nonsense.
local
 
Posts: 295
Joined: Mon Aug 05, 2019 1:19 pm

Next

Return to Sci.Physics.Foundations

Who is online

Users browsing this forum: No registered users and 73 guests

cron
CodeCogs - An Open Source Scientific Library