How Much Has Quantum Computing Actually Advanced? (ieee.org) 13
For a measured perspective on how much quantum computing is actually advancing as a field, IEEE Spectrum spoke with John Martinis, a professor of physics at the University of California, Santa Barbara, and the former chief architect of Google's Sycamore. From a report: IEEE Spectrum: So it's been about two years since you unveiled results from Sycamore. In the last few weeks, we've seen announcements of a 127-qubit chip from IBM and a 256-qubit neutral atom quantum computer from QuEra. What kind of progress would you say has actually been made?
John Martinis: Well, clearly, everyone's working hard to build a quantum computer. And it's great that there are all these systems people are working on. There's real progress. But if you go back to one of the points of the quantum supremacy experiment -- and something I've been talking about for a few years now -- one of the key requirements is gate errors. I think gate errors are way more important than the number of qubits at this time. It's nice to show that you can make a lot of qubits, but if you don't make them well enough, it's less clear what the advance is. In the long run, if you want to do a complex quantum computation, say with error correction, you need way below 1% gate errors. So it's great that people are building larger systems, but it would be even more important to see data on how well the qubits are working. In this regard, I am impressed with the group in China who reproduced the quantum supremacy results, where they show that they can operate their system well with low errors.
John Martinis: Well, clearly, everyone's working hard to build a quantum computer. And it's great that there are all these systems people are working on. There's real progress. But if you go back to one of the points of the quantum supremacy experiment -- and something I've been talking about for a few years now -- one of the key requirements is gate errors. I think gate errors are way more important than the number of qubits at this time. It's nice to show that you can make a lot of qubits, but if you don't make them well enough, it's less clear what the advance is. In the long run, if you want to do a complex quantum computation, say with error correction, you need way below 1% gate errors. So it's great that people are building larger systems, but it would be even more important to see data on how well the qubits are working. In this regard, I am impressed with the group in China who reproduced the quantum supremacy results, where they show that they can operate their system well with low errors.
Quantum computing might not do anything (Score:2)
It's still quite possible that the energy or storage cost of maintaining coherence among N qubits is exponential in N. This would make quantum computing merely a simulation curiosity, because it would be useless for encryption cracking.
Re: (Score:2)
Sure encryption cracking is not the only (or even the primary) purpose to which people wish to apply quantum computing?
Re: (Score:2)
Sure encryption cracking is not the only (or even the primary) purpose to which people wish to apply quantum computing?
It is one of the few things it can to well if it ever works. It is completely unusable for most things computers do.
Re: Quantum computing might not do anything (Score:2)
And it might, at the same time. You have to measure to know.
Re: (Score:2)
It's still quite possible that the energy or storage cost of maintaining coherence among N qubits is exponential in N.
There have been significant advances in reducing the size and increasing the reliability. They way I see it, we've built the equivalent of 1940s computer technology and radical changes are coming forth when someone invents a game-changer analogous to the invention of the transistor.
Re: (Score:2)
Re: (Score:2)
It's still quite possible that the energy or storage cost of maintaining coherence among N qubits is exponential in N. This would make quantum computing merely a simulation curiosity, because it would be useless for encryption cracking.
Indeed. Or that you actually do not get the reliability needed to run long, complex calculations without far too many retries needed.
Re: (Score:2)
If that would be the case, it would be relevant basic research.
An unexpected scaling in the decoherence of Quantum Error Correction Algorithms would falsify the current understanding of quantum mechanics.
If Quantum computation does not scale (after solving the technical problems), that could be more relevant scientifically than Quantum Computers working.
So many papers and nothing real. (Score:2)
The bottom line is your are going to lose.
Re:So many papers and nothing real. (Score:4, Insightful)
> Quantum computing is PHd's gambling.
Note that it is EXACTLY what we want researchers to do. Many of them are paid with public money. We want them to go for low chance of success but major social impact if successful, because no one else will. The high chance of success low-impact stuff, that's what the every day companies are for.
With Quantum computing we are transitioning to a place where it is no longer a pipe dream, we got to the realm of "okay, maybe that can work". That's why you see major companies going for it. The theoretical ground work has been laid in the 1990's and early 2000. And they have been trying to build a practical machine since.
I no longer disbelieve that quantum computer (that can do something useful) might happen in my lifetime.
There isn't any actual quantum computing yet (Score:2)
All of the advances, so far, are in fabrication and error correction. I think of it a bit like fusion reactor research. There is still an enormous amount of work to do, just to find we don't know what we don't know. Having actual quantum computing used in a production environment, to solve real computing problems, is still very far away.
Proof is in the pudding (Score:2)
where they show that they can operate their system well with low errors.
The experiment doesn't, and can't, show that. All it does is output a random number using an algorithm that is slow on conventional computers. There is no opportunity to test your gate errors in that experiment. It is just a "self simulator" where you turn it on, and see what what falls out, where there can be no wrong answer.
Making the same mistakes as in AI (Score:2)
The world was assured that human-level AI was around the corner in the late 60s. It did not happen. It still hasn't happened. We do have AIs that excel at very narrow, very constrained domains, which fail spectacularly when the parameters change every so slightly.
The quantum computing world seems to be following the same route. Those frequent press announcements about breakthroughs - which, when examined in detail turn out to incremental, and often minute ones - are pushing quantum computing to its AI winte