Advanced quantum computer made available to the public for first time

Alex Wilkins

A quantum computer has solved a task in 36 microseconds that would take the best computer at least 9000 years to complete. It is the first time that a quantum computer has been made available to the public.

The strange properties of quantum mechanics allow quantum computers to perform calculations much faster than conventional computers. The goal in the field is to demonstrate that quantum computers can beat regular machines. The problem of sampling random numbers that is impossible for classical machines can be solved with the help of the Sycamore processor, which was the first to do so in 2019.

Jonathan Lavoie and his colleagues at Xanadu Quantum Technologies in Toronto, Canada, have built a quantum computer called Borealis that uses particles of light to solve a problem known as boson sampling. This involves measuring the properties of a large group of entangled, or quantum-linked, photons.

Boson sampling is difficult for ordinary computers because of the high complexity of the calculations. Borealis directly measures the behavior of up to 216 entangled photons.

It isn't particularly useful to solve this problem, but it is an important test.

The second device to demonstrate quantum advantage is Borealis. Researchers at the University of Science and Technology of China created the first machine. In 2020, it showed quantum advantage with 76 photons, and then again in an improved version in 2021. Last year, the USTC team demonstrated quantum advantage in the random-number-sampling problem with a machine called Zuchongzhi.

More power

Peter Knight at Imperial College London says that the Chinese experiment was a tour de force and that the system called Borealis is an advance on it.

Compared with Borealis, Jiuzhang uses a larger number of beam splitters. Borealis uses loops of optical fibre to delay the passage of some photons relative to others, instead of space.

The stripped-back design ofBorealis allows it to be reprogrammed for people to run it on their own.

Knight says that people will probably begin by testing variations of boson sampling, but later on it may be possible to apply Borealis to different problems. No one has been able to demonstrate quantum advantage for a random-sampling problem that was tackled by Google.

Raj Patel at the University of Oxford says that the jump forward in scale over Jiuzhang is impressive, but it falls short of being a quantum computer. An interferometer, which measures interference patterns to extract information from the photons, has been limited to only recording certain photon interactions in an effort to get clearer readings.

Lavoie and his colleagues are working on a plan to build a fault tolerant, integrated chip that would improve the quantum machine's capabilities even further.

The journal Nature has a reference.