7 ms·
The recent “How Quantum is the D-Wave Machine?” Shin et.al. paper
- nmc 13y agoImpressive. This guy goes to the extent of linking to the transcript of a talk perfectly irrelevant to the point. But he cannot link to the paper he is criticizing. So here it is: Shin et al. How "Quantum" is the D-Wave Machine? http://arxiv.org/abs/1401.7087 http://arxiv.org/abs/1401.7087
- ot 13y agoScott Aaronson is one of the main D-Wave skeptics. His most recent post on the subject has a nice overview on the latest works on d-wave "quantumness": http://www.scottaaronson.com/blog/?p=1643 http://www.scottaaronson.com/blog/?p=1643
- Guvante 13y agoWasn't one of the major points about the D-Wave machine that while it appeared to work quantum like with low numbers of qubits, it appeared classical once you actually started doing real work? This would perfectly align with the defended studies. Additionally it would also align with what skeptics claim it is doing, performing some efficient classical analysis on custom hardware. Unfortunately the meat is missing from this response, specifically a non-trivial example of definitely quantum data.
- kposehn 13y ago"When I saw how trivially wrong it was it was like opening a Christmas present and getting socks." Wait, what is wrong with getting socks at Christmas? >_>
- Tloewald 13y agoHe can't use his ju-jitsu while wearing socks!
- higherpurpose 13y agoI think the jury will remain out on this one for the next few generations of D-wave computers. If by the time it reaches 2k-4k qubits, they still can't prove without a doubt that it's at least a special form of quantum computing, then I guess then we'll have our final answer.
- cdibona 13y agoFrom what we've seen (on the nasa/google installation) we're seeing quantum operation, but yes, once we see 2k qubit chips (which we hope happens within the year) that's when we'll know about whether there is an actual advantage vs scalar processors for annealing problems. This is one of the reasons we're all going round and round on proving 'quantumness' so that when we get a processor with more qubits we can predict and test that it is indeed all that an a bag of chips.
- levlandau 13y agohow do you know you are seeing quantum operation?
- lowglow 13y agoI recently had the pleasure of talking about this particular subject over beers with someone very close to the quantum computing community. The short answer is: "They aren't really doing quantum computing. They're essentially cheating." But, I'm no quantum mechanic PhD, so it's not first hand knowledge as to why.
- fidotron 13y agoIt seems quite likely that whoever turns quantum computing into something practical will face exactly that criticism from the academic quantum computing community, largely since it appears construction of what they think of as proper quantum computers doesn't seem to be too feasible. If D-Wave have managed it I must admit I'd be surprised, but the amount of venom thrown their way is curious.
- CJefferson 13y agoThe venom is understandable in any competition (and the battle to build a working, useful quantum computer is a major competition between both companies and universities) where just about every knowledgeable person is convinced someone cheated.
- TallGuyShort 13y ago>> chief technology officer of D-Wave and 2010 NAGA Brazilian jiu-jitsu light heavyweight world champion. He could always challenge skeptics in a fight to the death.
- troymc 13y agoHe starts with his philosophy of what science is, which is apparently based on a TED talk by David Deutsch: "...good explanations for why things are the way they are." Oh no. Science, emphatically, does not answer why, or at least not the ultimate why. It answers how; it provides a set of models. To quote Feynman, "While I am describing to you how Nature works, you won't understand why Nature works that way. But you see, nobody understands that." from QED: The Strange Theory of Light and Matter, p. 10.
- lutusp 13y agoThanks for making this important point. Science explains, it doesn't justify or rationalize. And each of science's explanations is open to falsification by new evidence.
- tbrownaw 13y agoActually, "how" is "why".
- Eliezer 13y agoSpeaking as a relatively qualified epistemologist, I nod.
- lutusp 13y ago> Actually, "how" is "why". In a scientific context, "how" and "why" are generally regarded as separate questions. Consider the classic case of peppered moth evolution: the "how" is answered by natural selection -- through random genetic variations the moths changed color over time to match their changing environment. The "why" is answered by noting that the camouflaged moths escaped predation by moth-eating species.
- darkmighty 13y agoI agree that It does not answer why in general, but that quote in particular seems reasonable -- Deutsch is referring to the whys which result from more fundamental interactions (more fundamental models); although eventually, you reach the basic laws of science and then there's really no point in further asking why.
- lutusp 13y ago> D-Wave answers "How Quantum is the D-Wave Machine?" This is the first time I've heard "quantum" used as an adjective. But probably not the last.
- bcbrown 13y agoIsn't the 'quantum' in "quantum mechanics" an adjective?
- lutusp 13y ago> Isn't the 'quantum' in "quantum mechanics" an adjective? Is "mathematical" in "mathematical physics" an adjective or a noun? I think it's a noun. On the other hand, the "extreme" in "extreme physics" is an adjective. Now for a gray-area example. If I say "very mathematical physics", then "mathematical" becomes an adjective. And I might be wrong about this. :)
- duskwuff 13y ago> I think it's a noun. On what grounds? It's being used to modify a noun ("physics"), and can be modified by an adverb ("very"). This is characteristic of an adjective.
- twic 13y agoHow quantum? Very quantum. Such entanglement. Wow.
- bestrapperalive 13y agoWhen I read this blog post, it was like opening a Christmas present and getting socks. I don't know much about the state of the art, the points in contention between the people in this argument, or quantum computing in general. An article that explained the competing standards for what constitutes a quantum computer and either made a case for why one standard is more useful or described how one participant's argument was inconsistent with their standard would have been genuinely helpful. Instead I read a link to a TED talk, epistemological meta-babble, and a promise that further experiments are incoming. This company may be completely legitimate and its product may be the most revolutionary thing since sliced bread, but I didn't know before I read this post, I don't know now, and I now think their CTO sounds too Steorn-y for comfort.
- foobarqux 13y ago"D-Wave's Dream Machine" is a good article I submitted recently that never made the front page. https://news.ycombinator.com/item?id=7128382 https://news.ycombinator.com/item?id=7128382