First, let us explain the headline for this article. At the Quantum Computing Report, we see a distinct difference between Quantum Supremacy and Quantum Advantage. Quantum Supremacy is defined as demonstrating that a quantum computer can solve a problem, even if it is a contrived problem with no real world value that could not possibly be solved on a classical computer within a reasonable time frame. Quantum Advantage is defined as solving a real world problem that has commercial value that cannot be solved on a classical computer. Although we see that achieving Quantum Supremacy is an important stepping stone on the road to Quantum Advantage, it should not be the end goal.
This week, a group from the University of Science and Technology of China, led by physicist Jian-Wei Pan, published a paper describing the achievement of Quantum Supremacy using a Gaussian boson sampling quantum computer. We had previously described last September the research that the USTC group has been performing and we had been expecting that such a paper with additional details on this research would be published soon.
The photonic technology that the group uses a collection of lasers, mirrors, prisms, and photon detectors can generate up to 76 of what they term output photon clicks (rather than qubits). In comparison, the Google superconducting quantum computer which provided the first demonstration of quantum supremacy had 53 qubits. So, the state space of the USTC machine exceeds that of Google’s by about many orders of magnitude.
But still the question remains whether this technology can be leveraged to demonstrate Quantum Advantage on a real world problem. At this point, no one really knows. The Gaussian boson sampling quantum computer cannot be regarded as fully programmable quantum computer and would not have the flexibility that could be offered the true gate based computers in development at Google, IBM, Rigetti, IonQ, Honeywell, and many others. In addition, there is no mention of a software programming environment available for a Quantum Boson Sampler, like there is with the other machines, and we deem this as a critical requirement to be able to reach Quantum Advantage. For more on this technology, you can view the article we previously published about this technology in the Quantum Computing Report, a technical analysis in our Qnalysis section, a commentary by Scott Aaronson, an article just published in Scientific American with a summary of the latest findings, and a full article in Science magazine that describes this technology in detail.
December 4, 2020