08 Oct 18
https://www.quantamagazine.org/graduate-student-solves-quantum-verification-problem-20181008/
The quantum verification problem now solved by her, how do you know a quantum computer is actually doing quantum stuff?
She worked out a protocol I guess sounds like a kind of Turning test for quantum computers that answers the question of a quantum computer actually doing quantum computing.
She is a rising star!
09 Oct 18
@sonhouse saidYou're not sceptial at all?
https://www.quantamagazine.org/graduate-student-solves-quantum-verification-problem-20181008/
The quantum verification problem now solved by her, how do you know a quantum computer is actually doing quantum stuff?
She worked out a protocol I guess sounds like a kind of Turning test for quantum computers that answers the question of a quantum computer actually doing quantum computing.
She is a rising star!
I would like to have confirmation from more independant places before I swollow this fantastic news.
@fabianfnas saidWell, I might be skeptical but not sceptial at all😉 I would imagine they have already vetted the process by which she proves a QC does quantum magic. I think there are enough QC chips out there to test her theory and that is probably why they accepted her paper even though QC chips at the present level are somewhere like computer chips were back in 1974, if you remember the 4004 chip, a 4 bit computer chip superseded by the 8008, the first 8 bit cpu and then the 8080 and that subsequent line of advancement we have now going from maybe 1 Mhz speed of those very early days to the multi gigahertz versions of today.
You're not sceptial at all?
I would like to have confirmation from more independant places before I swollow this fantastic news.
There are QC's out there right now but maybe good enough to add two and two which just adds evidence they are actually doing QC things which is what this paper is all about. My guess is she found something real.
09 Oct 18
1 edit
@sonhouse saidI had a look at her paper. The proof depends on a learning algorithm being quantum intractable. This implies to me that the question is already basically settled, since what she's ruled out is the quantum machine learning to produce the result the verifier expects rather than doing a "real" quantum computation.
Well, I might be skeptical but not sceptial at all😉 I would imagine they have already vetted the process by which she proves a QC does quantum magic. I think there are enough QC chips out there to test her theory and that is probably why they accepted her paper even though QC chips at the present level are somewhere like computer chips were back in 1974, if you remember t ...[text shortened]... tually doing QC things which is what this paper is all about. My guess is she found something real.
09 Oct 18
@deepthought saidWow, learning? Wouldn't that imply some kind of consciousness? How would it be able to 'learn'? How would it know what the operator wants?
I had a look at her paper. The proof depends on a learning algorithm being quantum intractable. This implies to me that the question is already basically settled, since what she's ruled out is the quantum machine learning to produce the result the verifier expects rather than doing a "real" quantum computation.
09 Oct 18
@sonhouse saidDon't get carried away by that, the learning algorithm just represents a way the machine could fool the verifier. They want to be able to prove the alleged quantum machine does quantum computation and doesn't arrive at its result by cheating in some way. The learning algorithm being tractable to the machine being tested would leave scope for a non-quantum algorithm to have solved the problem, undermining the claim that the machine is indeed a quantum computer. If the learning algorithm is computationally intractable even to the claimed capability of the machine then there is no alternative to the conclusion that the machine solved the problem directly by quantum computation.
Wow, learning? Wouldn't that imply some kind of consciousness? How would it be able to 'learn'? How would it know what the operator wants?
14 Oct 18
@deepthought saidDo you think her solution will be proven?
Don't get carried away by that, the learning algorithm just represents a way the machine could fool the verifier. They want to be able to prove the alleged quantum machine does quantum computation and doesn't arrive at its result by cheating in some way. The learning algorithm being tractable to the machine being tested would leave scope for a non-quantum algorithm to h ...[text shortened]... o alternative to the conclusion that the machine solved the problem directly by quantum computation.
15 Oct 18
1 edit
@sonhouse saidThe quantum verification problem now solved by her, how do you know a quantum computer is actually doing quantum stuff?
https://www.quantamagazine.org/graduate-student-solves-quantum-verification-problem-20181008/
The quantum verification problem now solved by her, how do you know a quantum computer is actually doing quantum stuff?
She worked out a protocol I guess sounds like a kind of Turning test for quantum computers that answers the question of a quantum computer actually doing quantum computing.
She is a rising star!
Jabba the Hutt had the answer, if the machine fails to perform, you get your droids to turn it upside down and turn hot steam jet's on its feet. No!...No!...No!!...Ahhhhh!!! 😀
24 Oct 18
@mchill saidIt does! And it does not! It all depends on the probability wave at hand...
The quantum verification problem now solved by her, how do you know a quantum computer is actually doing quantum stuff?