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

New silicon structure opens the gate to quantum computers

princeton.edu

Post reply on HN
Print view View original

31–40 of 59 posts

Re: New silicon structure opens the gate to quantum computers

#31

Post by lisper »

himom wrote: post #17
tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

There is no added Turing power to QC over CC. Turing complete is Turing complete. Think of QC as a coprocessor for certain classes of algorithms like prime factorization.

With exponential speedup. That's the big deal.

Re: New silicon structure opens the gate to quantum computers

#32

Post by soberhoff »

frik wrote: post #29
tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

Can the idea of quantum computing be simplified/abstracted to say instead of doing binary (0 1) calculations, it do does calculation on arbitrary base? E.g. base 16 (hex) or base 1024? If this is true, it would be easier to code, and compilers could spit out very efficient machine code.

What? How does a compiler care about the base of your machine code?

Re: New silicon structure opens the gate to quantum computers

#33

Post by IncRnd »

tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

This should answer your question.

http://math.nist.gov/quantum/zoo/

However, quantum computing will likely do nothing for you as a person other than make your life more difficult due to complexity changes for solving non-pqc algorithms.

Re: New silicon structure opens the gate to quantum computers

#34

Post by PeachPlum »

tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

Decreasing lead time in manufacturing has dramatically changed the world during your lifetime. There are lots of unsolved problems in computational logistics that look like tsp. We have whole conferences on heuristics to shorten their runtimes. e.g. [1] And there are more. Hundreds of PhDs will be working today on just that topic.

[1] http://iwi.econ.uni-hamburg.de/iccl/front.html

Re: New silicon structure opens the gate to quantum computers

#35

Post by jcoffland »

bsder wrote: post #16
NhanH wrote: post #13

Earlier quoted context omitted.

Quantum annealing can be used for optimization of functions. Shor’s algorithm of integer factorization would solve (and cause) a whole host of problems for us. But in general, quantum computer is not strictly a superset of improvement over Turing machine. It’s just different. And while it might reduces your database query time, unless you are already working at that abstraction it might probably be invisible changes…

> Shor’s algorithm of integer factorization would solve (and cause) a whole host of problems for us. I thought it was still an open question as to whether a quantum computer would actually be able to solve this class of problems or not. In addition, quantum computers have a great deal of noise--so quantum error correction seems to be a thing.

Shor’s algorithm can factorize integers on a quantum computer asymptoticly faster than on a classical computer but the algorithm requires at least a 128 qubit quantum computer to factor a 128 bit number and thereby break something like 128 bit RSA encryption. No one yet has, or is publicly admitting to having, a reliable 128 qubit device.

In some cases, error correction can be achieved simply by running the algorithm a few times.

Re: New silicon structure opens the gate to quantum computers

#36

Post by jcoffland »

tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

The traveling salesman problem is NP-complete and therefore not known to be solvable in less time on a quantum computer.

https://cstheory.stackexchange.com/questions/31084/travellin...

Re: New silicon structure opens the gate to quantum computers

#37

Post by kybernetikos »

tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

http://math.nist.gov/quantum/zoo/

Of particular note, there are significant speed ups for optimization and constraint satisfaction, cryptanalysis, machine learning and indeed 'looking up data in a database' (grover search).

Re: New silicon structure opens the gate to quantum computers

#38

Post by Sniffnoy »

ryacko wrote: post #30
Sniffnoy wrote: post #23

Earlier quoted context omitted.

I'm afraid you're a bit off. A quantum computer can search N possibilities (such as when, say, looking up data in a database, though I doubt you'd go to the trouble of using a QC for that) in time O(sqrt(N)). So for 2^N things, that's 2^(N/2), not 2^(sqrt(N)). Obviously still a big improvement.

But wouldn't it have to be stored as qubits? A database in qubits would be expensive.

Well, yes, that's one reason why you wouldn't actually use a QC for that...

Re: New silicon structure opens the gate to quantum computers

#39

Post by Sniffnoy »

frik wrote: post #29
tim-- wrote: post #10

My biggest pet peeve about quantum computing is that no one can answer me the question of "what can quantum computing do for me?". The answer I hear is that it helps solve the traveling salesman in record time, and that's all great and everything, but how will quantum computing be able to do things such as decrease the time it takes to train a RNN, or look up data in a database?

Can the idea of quantum computing be simplified/abstracted to say instead of doing binary (0 1) calculations, it do does calculation on arbitrary base? E.g. base 16 (hex) or base 1024? If this is true, it would be easier to code, and compilers could spit out very efficient machine code.

Yes, base has nothing to do with anything, talking about "qubits" is just a convention since computation is usually thought of in terms of bits. The fundamental idea of a quantum computer though has nothing to do with binary in particular. Same as for classical computers.

It's not clear to me why you claim this would be related to efficient machine code, though.

Re: New silicon structure opens the gate to quantum computers

#40

Post by techdragon »

badspeler wrote: post #21
propter_hoc wrote: post #9

I did graduate research in an area of experimental physics that was quite close to state of the art quantum computing research. It was very exciting, but all the research was pretty much in extreme conditions that aren't really usable for a realistic computing device (ultra-high vacuum, nanokelvin temperatures, high magnetic fields or laser trap confinement, etc.). From the looks of it this paper is in the same vein.…

That's true but once Science conquers something, engineering isn't very far behind. The apparatus to build Bose-Einstein Condensates was a fucking monster of vacuums and quantum optics but now we can fit a Bose-Einstein Condensate on a microchip. https://arxiv.org/abs/0912.0553

I Just read that paper and while It’s hardly just a microchip it’s damn impressive considering the technology involved and the requirements it’s fulfilling. It’s amazing that we can now build devices that can produce ultra high vacuums smaller than a hotel minibar, and design systems that are capable of active cooling down to <1K without the use of any cryogenic refrigeration at all.
Post reply on HN