Physicists from the Shanghai China University of Science and Technology (USTC) on their own quantum computer Jzhāng have implemented an experiment on sampling Gaussian bosons. The Jzhāng quantum system took 76 photons in 200 seconds, while a classical supercomputer would take 2.5 billion years.
An experiment at a Chinese university is not even a pen test. Scientists are still in doubt about the choice of physical principles on which quantum computers can be built, and are not even sure about the choice of the range of problems with which quantum systems will cope better than classical ones. Therefore, the experience gained, the path even it will not entail any practical results in the near future, is no less important than the obvious breakthrough.
At one time, theoretical physicists pinned many hopes on optical (photonic) quantum computers. This was due to the fact that photons promised to become qubits at room temperature, which would greatly simplify the assembly and operation of quantum systems. In practice, creating a multi-qubit optical quantum system would require millions of lasers and hundreds of millions of mirrors, prisms, and much more, which would make one forget about the notorious quantum superiority of optical quantum computers over classical ones.
The experience supplied by the Chinese has shown that even with a limited set of components, an optical quantum system is capable of surpassing the classical one, which means that there are prospects in this direction, you just have to work and work.
The problem of sampling Gaussian bosons by a classical computer is solved by the matrix method, which exponentially increases the computation time as the number of certain particles increases. An optical quantum computer is its own operating model for sampling Gaussian bosons. To calculate the result, you just need to experiment. And if each selection cycle on a quantum system takes up to 200 seconds, then the supercomputer must spend incomparably more time on this.
However, the Chinese installation presented by physicists is also not simple. The laser radiation is split into 25 beams and affects 25 crystals of potassium titanyl phosphate. After hitting each crystal, two photons fly out of it in opposite directions. The photons hit 100 entrances, each of which follows a path of 300 prisms and 75 mirrors. The system also has 100 exits with sensors at the end, which register particles that have reached the finish line. On average, over 200 second cycles, the USTC team detected about 43 photons per launch. But at one time they observed 76 photons – more than enough to justify a claim for quantum superiority.
A classical computer, under good circumstances, is able to calculate 50 photons in two days of calculations, and it may take 2.5 billion years to select 76 photons, as extrapolation says.
In addition to proving the fundamental possibility of achieving quantum superiority on an optical quantum computer, the result obtained can theoretically be applied to solving practical problems in the near future. For example, for solving specialized problems in quantum chemistry and mathematics. In a broader sense, the ability to manipulate photons as qubits is a prerequisite for the creation of a large-scale quantum Internet or its key elements in the future. In other words, in this case, the following thesis is appropriate “The road will be mastered by the walker”…
If you notice an error, select it with the mouse and press CTRL + ENTER.