“The University of Science and Technology of China is expected to achieve a 60-qubit, 99.5% fidelity superconducting quantum system this year, and in five years, it hopes to increase this number to 1,000 qubits.”
At the Mozi Salon, Zhu Xiaobo, a professor at the Shanghai Research Institute of the University of Science and Technology of China and the head of superconducting quantum computing in Pan Jianwei’s team, revealed the timetable for the development of quantum computers at the University of Science and Technology of China.
The 10-year goal is the same as Google, that is, a one-million-bit quantum computer with a fidelity of 99.8%, which is the same as Google.Although various quantum computer research and development forces frequently release hundreds or even thousands of qubits, Zhu Xiaobo believes that these statements only tell “one side of the story.”
It turns out that it is not difficult to increase the number of bits of a quantum computer, but it is necessary to ensure that each bit is accurately controlled. A very challenging key difficulty is: how to achieve scalability in the case of high-fidelity control and readout of single microwave photon states (qubits)
In his view, the more solid breakthroughs in recent years also represent the current international top-level results of superconducting quantum computing, which is Google’s 53-qubit, 99.4% fidelity superconducting quantum system. Zhu Xiaobo revealed that the 60-bit quantum computer that the China University of Science and Technology team will soon implement is just the benchmark, and it will make the processor three times more difficult to solve the same problem.
According to a paper published by Google in 2019, this processor named Sycamore solved the problem of the random circuit that the world’s most powerful supercomputer “apex” needs to run for 10,000 years in 200 seconds, and successfully achieved the so-called “quantum superiority” which means that the ability to deal with specific problems exceeds the current most powerful supercomputer.
In 2017, Professors Pan Jianwei, Lu Chaoyang, Zhu Xiaobo of the University of Science and Technology of China, together with the research group of Professor Wang Haohua of Zhejiang University, announced the use of high-quality quantum dot single-photon sources to build a multi-photon programmable quantum computing prototype for Bose sampling. After the manipulation of nine superconducting qubits publicly reported by Google, NASA, and UCSB, the entanglement of ten superconducting qubits was realized for the first time, and on this basis, a quantum algorithm for quickly solving linear equations was realized.
However, whether it is the random line of Google or the Bose sampling of the University of Science and Technology of China, these problems used to demonstrate the capabilities of quantum computers do not have practical application capabilities. Zhu Xiaobo hopes that within 10-15 years, quantum computers can solve “real problems” in fields such as cryptography.
He is also optimistic about the potential of quantum computers in the field of artificial intelligence. “Quantum machine learning is now a very popular direction. There are still problems in terms of input and output, but the processing flow is very advantageous.” As long as the number of bits of a quantum computer reaches a few hundred and the accuracy reaches 99%, it can overwhelm classical computers.
However, Zhu Xiaobo believes that when the day is truly practical, quantum computers will also appear more in the form of servers, and everyone will upload problems to the cloud and hand them over to the server for calculation.