5 ms·
Quantum Computing Is Real, and D-Wave Just Open-Sourced It
- sigmaprimus 10y agoI'm wondering why people think that the argument that because the Dwave system can only solve 1 problem or in fact only 1 subset of a problem disqualifies it from being a quantum computer is valid. Last time I checked a conventional computer only solves 1 problem at its core. Are quantum computers measured with a different stick?
- sigmaprimus 10y agoAlso who made the decision that a quantum computer has to be faster, or better or even solve harder problems?
- ThePhysicist 10y agoBut what problem can it actually solve?
- sigmaprimus 10y agoWell there is the salesman example in the article for one, I find the checker board problem much easier to wrap my head around. On a checker board you have 64 squares, half are red and half are black, everywhere a red touches a black square is a junction point. How many junction points are there? Now that you know how many junctions there are how many different ways can you lay out those red and black squares to make that same amount of junctions? What if 3/4 of the squares were black? My understanding is that the Dwave system will produce a number of answers that are close to correct, and a majority that are correct (probably). Things start getting really tricky when you make the squares into cubes and run the problem again, or if you want to get really bent add a fourth dimension to the problem.
- ThePhysicist 10y agoYes but even for the traveling salesman problem (which in most variants is NP-complete) there is -to my knowledge- no experimental or even theoretical proof that an adiabatic quantum computer could outperform a classical one. The fact that the D-Wave computer is a computationally complex quantum system does not automatically imply that the dynamics of the system can be harnessed to solve computationally complex problems, because we usually don't have sufficient control over the system. Given the ability to control individual degrees of freedom of a given system with arbitrary precision almost anything can be turned into an exponentially powerful classical or quantum computer, even a simple hot cup of tea. The problem is that we often don't have the means to exert the required control over the system, and this is the core problem with all current approaches to quantum computing, adiabatic or not. So, yes, you can encode the solution to many problems in the ground state of the Hamiltonian of a coupled qubit system, but reliably keeping the system in this ground state while transforming it into the target state turned out to be impossible in any real-world system so far, as even the tiniest thermal fluctuations will trigger state transitions during the annealing process and ruin the whole thing. So if D-Wave really claims they have a working quantum annealer on their hands I want to see proof, and if they can muster it I'm pretty sure they'll get a nobel prize in return. Show me the data!
- sigmaprimus 10y agoWow, blind me with brilliance or baffle me with bs, you are obviously much smarter than me and I conside to you. All that I ask is that you use your knowledge to help what I feel is a pioneering company with an idea that even the smartest minds...including yours have had to concede that these guys have something. They have pulled up their skirt and offered you an opportunity to make something great and you want to bicker over their har color.
- sigmaprimus 10y agoI must apologize for that last comment as I phrased it in a demeaning context..the truth is we (humanity) need your help and rather than fight over Synaptics I emplore you to focus your valuable experience and knowledge to pursue this area of computer science in the direction of achieving the greater good rather than a geekisH slap fight that gets us nowhere. You seem like a smart guy, fix the issue by creating something great and make america great again.
- ramshanker 10y agoTL:DR They open sourced a software library for encouraging "Not familiar with quantum physics" programmes to get started.
- sauronlord 10y agoThey still can't get their gate fidelity anywhere close. Nice PR move ...again
- Bombthecat 10y agoHeh, I would be really surprised if it would work "now". Maybe in five or ten years though
- ThePhysicist 10y agoIf it is real now, why can't we do the following: 1. Take a problem that is impossible to solve with classical computers (e.g. from combinatorics, number factorization), but whose solution can be verified easily. 2. Solve it on their quantum computer. D-Wave keeps saying they have a quantum computer since 5 years now, yet they have not produced any proof of quantum speed up achieved with their machine.
- wolfgke 10y agoFor a long time it was controversial whether D-Wave's systems even used quantum effects - but now the consensus is that they do. D-Wave's systems implement some (restricted) kind of adiabatic quantum computing/quantum annealing. Even from a theoretical perspective it is up to now hardly understood in what sense quantum annealing is more powerful (in terms of better possible running time opposed to classical annealing combined with specialized algorithms). So you cannot simply implement some well-known quantum algorithm as Shor's on D-Wave's systems.
- ThePhysicist 10y agoYes I understand that quantum annealing is different, but the whole point of having a quantum computer is that it is able to solve interesting mathematical problems, or at the very least is able to simulate other quantum systems efficiently (beyond nearest-neighbor Ising models). To my understanding, the D-Wave system has demonstrated neither of these things so far.
- wolfgke 10y ago> but the whole point of having a quantum computer is that it is able to solve interesting mathematical problems > To my understanding, the D-Wave system has demonstrated neither of these things so far. There are different opinions whether D-Wave's quantum computer is able to solve interesting mathematical problems. Here are some - decide for yourself whether you consider them as interesting or not (I'm personally on the side that it is not the kind of application I am interested in, but can imagine that there might be people for whose these might be useful): > https://www.quora.com/What-are-some-cool-possible-applications-of-the-D-Wave-quantum-computer https://www.quora.com/What-are-some-cool-possible-applicatio... Here D-Wave's marketing about applications - read them with a grain of salt ;-) : > http://www.dwavesys.com/quantum-computing/applications http://www.dwavesys.com/quantum-computing/applications