Computing News, Blogs
Within the last couple months, there was a major milestone in the quest to build a scalable quantum computer, and also a major milestone in the quest to figure out what you would do with a quantum computer if you had one. As I’ve admitted many times, neither of those two quests is really the reason why I got into quantum computing—I’m one of the people who would still want to study this field, even if there were no serious prospect either of building a quantum computer or of doing anything useful with it for a thousand years—but for some reason that I don’t fully understand, both of those goals do seem to excite other people.
So, OK, the experimental breakthrough was the Martinis group’s use of quantum error-correction with superconducting qubits, to preserve a logical bit for several times longer than the underlying physical qubits survived for. Shortly before this came out, I heard Krysta Svore give a talk at Yale in which she argued that preserving a logical qubit for longer than the physical qubits was the next experimental milestone (the fourth, out of seven she listed) along the way to a scalable, fault-tolerant quantum computer. Well, it looks like that milestone may have been crossed. (update: I’ve since learned from Graeme Smith, in the comments section, that the milestone crossed should really be considered the “3.5th, ” since even though quantum error-correction was used, the information that was being protected was classical. I also learned from commenter Jacob that the seven milestones Krysta listed came from a Science paper by Schoelkopf and Devorret. She cited the paper; the forgetfulness was entirely mine.)
In more detail, the Martinis group used a linear array of 9 qubits: 5 data qubits interleaved with 4 measurement qubits. The authors describe this setup as a “precursor” to Kitaev’s surface code (which would involve a 2-dimensional array). They report that, after 8 cycles of error detection and correction, they were able to suppress the effective error rate compared to the physical qubits by a factor of 8.5. They also use quantum state tomography to verify that their qubits were indeed in entangled states as they did this.
Of course, this is not yet a demonstration of any nontrivial fault-tolerant computation, let alone of scaling such a computation up to where it’s hard to simulate with a classical computer. But it pretty clearly lies along the “critical path” to that.
What is the most liberal startup news blog?
Save your draft before refreshing this page.