The true hardness of quantum error mitigation
Speaker: YihuiTitle: The true hardness of quantum error mitigation
Date: 27 Feb 2023 16:30-18:00 EST
Location: SEC Room 1.412
Food: Thai
Quantum error mitigation has been proposed as a `near-term’ surrogate to quantum error correction. Requiring no or few additional quantum resources, the premise of quantum error mitigation is that you can remove quantum noise from the result of a calculation using only classical post-processing. Algorithms for this are being intensely developed as we speak, but there are heuristic signs that they don’t scale, often requiring exorbitant sample complexities. We put these observations on solid theoretical ground by constructing rapidly mixing quantum circuits that show that exponential-in-system-size sample complexities are needed for error mitigation. Our results imply that, while improvements in quantum hardware will push noise levels down, if error mitigation is used for certain near-term applications, ultimately this can only lead to an exponential time algorithm with a better exponent when compared with classical algorithms, putting up a strong obstruction to the hope for exponential quantum speedups in this setting.