DYNA JOURNAL ENGINEERING

Jian-Wei Pan and Chao-Yang Lu, both from China University of Science and Technology, and their colleagues improved a quantum computation technique called "boson sampling" to achieve a 14-photon detection record in their final results. Previous experiments were limited to just five photons. The increase in the number of particles is small, but it is equivalent to a gain of 6.5 billion times in "state space," or the number of ways a computer system can be configured. The larger this state space, the less likely it is that a classical computer can do the same calculation.

In classical computers, information is encoded in binary bits, so two bits can be 00, 01, 10, or 11. A quantum computer can be in all classical states simultaneously: two qubits have some probability of being 00, 01, 10, and 11 until they are measured; three qubits have a probability of being in any of the eight states; and so on. This exponential increase in information explains why quantum computers have such an advantage in theory.

We have to recognize that this experiment has a limited utility. A universal computer can solve any kind of problem and with this system we can only solve one. But solving a single problem faster than a classical computer would be a demonstration of quantum-computational supremacy.

Real problem is time. The team needs to produce individual photons separately and simultaneously. Photons do not wait for each other, so it is necessary to generate them with great precision.

If the photons arrive even with a few trillionths of a second difference, they are "lost". Each photon in the system increases the possibility that there are photons out of synchronization, because the error will aggravate the error. The more photons that are lost, the easier it is for a classical computer to simulate the distribution of photons, and the further one goes from quantum-computational supremacy.

While the researchers detected only 14 of the 20 incoming photons, that number was enough to generate a hard-to-calculate state space.