Borealis Quantum Computer (Photo: Xanadu Quantum Technologies Inc.)
Quantum computers that are still in experiments are based on quantum bits, but can also use photons, each particle of light. Now scientists have taken new steps in this latest model and achieved a task in just 36 microseconds that would take the classics almost 9000 years.
The description of this quantum photon processor is called BorealisPublished in the journal Nature and its managers believe This is a photonic experiment with the largest quantum advantage, showing their improvement over classical systems..
“On average, the best-available algorithms and supercomputers will take more than 9,000 years to complete,” said researchers from Xanadu, the Canadian company for quantum technology, and the US National Institute of Standards and Technology.
This system represents improvements over previously demonstrated photonic devices and could be an important step towards the development of quantum computers, concludes Jonathan Lavois’s scientific team.
One of the major drawbacks of quantum devices – both qubits and photon-based devices – is that they outperform the classical systems, computers, and supercomputers currently on the market that establish quantum advantage or dominance.
But to date only a small number of experiments have shown this achievement, especially in models based on quantum bits – including contrast when Google said it had reached quantum supremacy in 2019, which IBM questioned..
Now what has been published is to demonstrate this advantage in the processor with photons and the approach to demonstrating it is called the boson pattern, the photon is an example of a boson, an elementary particle.
This sampling is a calculation performed on a circle through which photons move, With a series of entrances and exits and a network of mirrors and fixed lensesWith other quantum optical instruments.
In fact, the calculation is based on certain parameters to determine how many photons end up due to changes inside the circuit, in a given output band and not in another.
Archive image of a Blue Gene Q supercomputer during a press screening at the Federal Polytechnic School of Lausanne (EPFL). EFE / Jean-Christophe Bott
And that is that the circle, as the researcher Carlos Sabin explained, from the Department of Theoretical Physics of the Autonomous University of Madrid, consists of a series of transformations that will be carried out on everything that is included in it.
These transformations can be caused, for example, by the splitting of a beam – an instrument that splits a beam of light in two – causing a certain probability that the photons will change strips in a circle and achieve their redistribution at the exit.
It seems silly, says Sabin, who is not involved in this study, but is not; Years ago it was shown that this calculation – knowing how many photons are in a given output line – is quickly impossible for ordinary computers.
And this is that there is a limit to photons above which classical computers can not perform calculations in a reasonable amount of time.
“If the circuit parameters are selected randomly and from a certain number of particles and from the input and output bands, it is almost impossible to calculate the probabilities with respect to the output for a conventional computer,” the researcher sums up Efe.
In the Nature Study, the team achieved the largest sample of the boson to date, with 216 stripes (average 125 photons) and calculated in record time: 0.000036 seconds.
While these claims are sometimes questioned (there may be better methods of computer computation than the authors suggest), these numbers go beyond previous boson sampling experiments and quantum dominance experiments in bits. “Quantum superconductors from Google.”
“The results must be embedded in the race to demonstrate quantum supremacy,” concludes the physicist, who emphasizes that the system can easily be programmed to generate certain states “by which it is known that universal quantum computing can be performed.”
This, he adds, responds to the most common critique of Boss’s selection “by which it is known that universal quantum computing can be performed”: its practical uselessness beyond the demonstration of quantum supremacy.
(Reported by EFE)
Continue reading: