Researchers Tao Wang and Yun Shang at Chinese Academy of Sciences, have developed a new error mitigation technique to improve the reliability of calculations performed on contemporary quantum computers. Their work introduces a hybrid Gaussian-exponential extrapolation scheme specifically tailored for quantum circuits exhibiting periodic structure, a common characteristic of many quantum algorithms. By modelling noise amplification using a log-normal distribution and incorporating Gaussian variance corrections, the method requires minimal prior noise characterisation and demonstrably reduces bias in calculations across a range of circuit types, including Trotterized Ising dynamics, random circuits, and Grover search. This advancement represents a key step towards obtaining more accurate results from near-term quantum hardware, which is currently limited by the inherent fragility of quantum states and susceptibility to environmental disturbances.
To read more, click here.