Thursday, December 8, 2016

survey of the qubit landscape

Last week, I poked fun at this Science infographic regarding different kinds of quantum bits (qubits), insofar as topological qubits are as good as unicorns at quantum computing. You could actually replace 'topological qubits' with 'transmogrifier' and the metrics wouldn't change. However, I’d like to highlight that the infographic (and corresponding article) is a pretty good summary of the state of affairs in quantum computing, so I’ve reproduced it here:

Infographic summarizing certain figures of merit for different implementations of quantum computing. Although trapped ions are shown to have a coherence time over 7 orders of magnitude higher than superconducting qubits, the qubit operations for the superconducting style are much, much faster as well. One day I'll have to assemble number of operations per coherence time for comparison. [(GRAPHIC) C. Bickel/Science; (DATA) Gabriel Popkin]

The various styles of quantum computing are told from the lens of ionQ, a hybrid university/startup founded by University of Maryland professor Chris Monroe. Monroe, along with his former boss Dave Wineland of NIST Boulder, is best known for achieving single-atom quantum control in 1995, for which Wineland won the Nobel Prize in 2012, the only such prize in the nascent field of quantum computing. Using ion traps for such atomic control was long thought of the most promising route to building a quantum computer, but recent progress has been hindered by an unexplained heating problem and quadratic scaling complexity. In a magnificent play on words, the article explains that "[trapped] ions are 'a bit of a black sheep right now,' Monroe admits, 'but I think in the coming years people will be flocking to it.'" And although the field is speckled with once-promising techniques such as NMR and optical quantum computing, I don't have much reason to doubt him.

Recently, industry has entered the fray of quantum computing, and superconducting qubits [on which I work] are the style favored, to the point where superconducting qubits and ions have been developed roughly to the same level. There are still some holdouts, most notably topological qubits at Microsoft and quantum dots/silicon impurities in Australia. And the startups are mostly coming from university collaborations helmed by experts in the field, as ionQ with Maryland and Quantum Circuits with Yale, although rare venture capital funded companies such as Rigetti Quantum Computing are out there as well (the latter two of which are using superconducting qubits). While the use of hybrid quantum systems is likely what it takes to build a working quantum computer, in the near future, research groups will continue to advance their favorite style and justify why it's the best system.

No comments:

Post a Comment