A computer capable of quantum supremacy — proving supremacy over conventional machines — is the first computer anyone can use over the internet
A quantum computer that encodes information with pulses of light accomplishes a task in 36 microseconds that would take the best supercomputers at least 9,000 years to complete. The researchers behind the machine also connected it to the internet allowing others to program it for their own use – For the first time such a powerful quantum computer has been made available to the public.
Quantum computers rely on the strange properties of quantum mechanics to theoretically perform certain calculations faster than conventional computers. A long-term goal in the field of quantum supremacy or quantum supremacy is to demonstrate that quantum computers can Actually beats normal machines. Google pioneered doing so in 2019 with its Sycamore processor which can solve problems involving sampling random numbers that are largely impossible with classical machines.
Now Jonathan Lavoie of Xanadu Quantum Technologies in Toronto Canada and his colleagues have built a quantum computer called Borealis that uses particles of light or photons traversing a series of fiber-optic loops to solve a problem called boson sampling . This involves measuring Properties of a large group of entangled or quantum connected photons separated by a beam splitter.
Boson sampling is a daunting task for ordinary computers because the computational complexity rises dramatically as the number of photons increases. Borealis essentially calculates the answer by directly measuring the behavior of up to 216 entangled photons.
Solving this question isn’t particularly useful beyond determining that quantum supremacy has been achieved but it’s an important test. “By demonstrating these results using Borealis we have validated key technologies required for future quantum computers” the way.
Borealis is the second device to demonstrate quantum superiority in boson sampling. The first is a machine called Jiuzhang created by researchers at the University of Science and Technology of China (USTC). It first showed quantum superiority with 76 photons in 2020 then again with an improved version 113 photons are used in 2021. Last year the USTC team also demonstrated quantum superiority in the random number sampling problem with a machine called Zu Chongzhi.
Peter Knight of Imperial College London said Borealis was an advance on Chapter Nine because it was a more powerful system capable of computing more photons and had a simplified architecture. “We all think the Chinese experiment is a stunt but we can’t Seeing it goes even further because there’s a limit to how much you can cram into an optical table,” he said.
Compared to Borealis Jiuzhang uses more beam splitters to send entangled photons in many different directions. But Borealis took a different approach using a fiber optic ring to delay the passage of some photons relative to others – separating them in time instead of space.
Another benefit of the stripped design is that the computer is easier to control so it can also be reprogrammed remotely letting people run it with their own settings. “Borealis is the first machine with quantum computing benefits that anyone can use Internet connection,” Lavoie said.
One might start by testing for changes in boson sampling Knight said but later might apply Borealis to a different problem. So far no one has been able to demonstrate quantum superiority for “useful” computing tasks – first random sampling problems The problem Google solves has basically no applications other than demonstrating quantum superiority.
Raj Patel of the University of Oxford said that while Borealis is an impressive leap in scale it’s not quite a fully programmable quantum computer like Sycamore or Zuchongzhi. This is because components called interferometers measure interference patterns to Extracting information from photons is limited to recording certain photon interactions for clearer readings. “To create a machine that’s programmable and can solve real-world problems you really want the interferometer to be fully connected,” Patel said.
Lavoie and his colleagues are now working to turn the blueprint they published last year into a scalable fault-tolerant photonic processor based on integrated chips which will further improve the capabilities of quantum machines.