Making quantum error mitigation work - insights from theory and experiments

Current quantum computers are on the one hand too small and noisy for implementing quantum error correction codes, and on the other hand not reliable enough for running calculations without error correction codes. Quantum error mitigation (QEM) is an emerging new approach that aims to generate reliable outputs using the currently available quantum computers. QEM methods can substantially mitigate the noise by running additional measurements that extract information about the noise. Unfortunately, these methods are either non-scalable, or heuristic and valid only for a specific nonrealistic noise model. We present a QEM technique we call KIK which is derived from a rather general Master equation. This method can handle a broad range of time-dependent noise mechanisms even when their amplitude is considerably strong. In the first stage, we intend to apply our method for ground state calculations of molecules in quantum computers and for producing better calibration procedures for quantum gates. Finally, a spin off of our method can be used to efficiently measure the quantum fidelity of a circuit without knowing what is the ideal output state.