## Recycled photons set fresh quantum computing record

来源：未知 作者：壤驷堵 时间：2019-03-14 01:11:00

By Jacob Aron Reusing old computer parts sounds like a terrible way to boost processing power, but it has enabled a quantum computer to set a new algorithmic record. Anthony Laing and colleagues at the University of Bristol, UK, have used recycled quantum bits, or qubits, to carry out a quantum calculation known as Shor’s algorithm on a larger number than ever before. The algorithm exploits quantum mechanics to simplify the factorisation of numbers into their prime components – a hard task for ordinary, classical computers when the numbers get really large. However, until now, the largest number factorised using Shor’s algorithm was 15. “There have been four demonstrations of 15 in the last 10 years, that’s as good as we’ve got,” says Laing. Part of the difficulty in reaching higher numbers is creating enough qubits to do the job. Laing and colleagues used a photon as their qubit. They realised it was possible to split Shor’s algorithm into parts and run one part at a time on a single photon. This slows things down, but for huge numbers it should still be much faster than a classical computer. The team factorised the number 21, a record for Shor’s algorithm but far from the large numbers needed if quantum computers are to outperform their classical counterparts on the task. An alternative quantum algorithm was used to factorise 143 earlier this year, having previously achieved 21. This algorithm uses a method called adiabatic quantum computing that, unlike Shor’s algorithm, is not mathematically guaranteed to provide faster performance for larger numbers. Laing hopes his recycling technique will more reliably help quantum computers scale up to a point at which they can carry out calculations that are inaccessible to classical computers. “Recycling should enable larger demonstrations more rapidly than has been realised before,” he says. There could be a downside to this advance, too: the ability to factorise numbers with more than about 300 decimal digits would be deadly for cryptography techniques widely used today, many of which rely on the difficulty of factorising large numbers. However, Liang says this is still a concern for the far future. “You don’t have to worry just yet; it’s a long way off.” “I’d call qubit recycling a clever ‘software’ technique, which allows us to get the most out of current experiments,” says Dan Browne at University College London, who has previously worked on implementing Shor’s algorithm. Attaining really big improvements will require scaling up quantum hardware too, he says. Laing’s team is also working towards that end. As part of the recycling technique, the researchers chained together two logic gates – a first for an optical quantum computer and an essential step in terms of hardware for building more complex devices. “To me, this achievement is more important than factoring 21, although to do so is an excellent demonstration of the technique’s power,” says Browne. Journal reference: Nature Photonics, doi.org/jkf More on these topics: