Qubits are inherently noisy, and even more so with current noisy-intermediate quantum devices. Several factors contribute to noise and interfere with the accuracy of quantum calculations. The more qubits that are needed for a particular calculation, the noisier it will be.
Pacific Northwest National Laboratory scientists Bo Peng and Niri Govind, along with Yuri Alexeev and Sahil Gulania from Argonne National Laboratory (ANL), discovered a way to perform more accurate quantum time dynamics of model systems on noisy quantum circuits.
“The potential for quantum computers to surpass classical ones is there,” said Peng who is the first author of the paper published in Physical Review A. “But this quantum supremacy just won’t happen unless we can get around the noise problem.”
Shallower circuits, reduced noise
Much like classical computers, quantum computers use circuits to run algorithms.
When a scientist wants to solve a computational problem, they need to run an algorithm that describes their problem on the computer. More complex algorithms require more extensive circuits to run. Quantum computers get noisier as the circuits get more extensive.
With their collaborators from ANL, Peng and Govind devised a way to run the same algorithms using shallower circuits. By exploiting symmetries revealed by the Yang-Baxter equation, they re-represented the quantum circuits to compress them efficiently.
Govind and Peng then tested their circuit compression on a real quantum computer: IBM’s Manila device. The results were stunning. “Not only could the compressed circuit run the simulation we gave it, but it also performed much better than the same uncompressed circuit,” said Govind. “Though more research is necessary, this shows a promising first step towards bypassing noise in quantum circuits.”
This research was supported by Q-NEXT, a Department of Energy National Quantum Information Science Research Center led by ANL.