Thrifty shadow estimation: re-using quantum circuits and bounding tails

Abstract

Randomized shadow estimation is a recent protocol that allows estimating exponentially many expectation values of a quantum state from “classical shadows”, obtained by applying random quantum circuits and computational basis measurements. In this paper we study the statistical efficiency of this approach in light of near-term quantum computing. In particular, we propose and analyze a more practically-implementable variant of the protocol, thrifty shadow estimation, in which quantum circuits are reused many times instead of having to be freshly generated for each measurement (as in the original protocol). We show that the effect of this reuse strongly depends on the family of quantum circuits that is chosen. In particular, it is maximally effective when sampling Haar random unitaries, and maximally ineffective when sampling Clifford circuits (even though the Clifford group forms a three-design). To interpolate between these two extremes, we provide an efficiently simulable family of quantum circuits inspired by a recent construction of approximate t-designs. Finally we consider tail bounds for shadow estimation and discuss when median-of-means estimation can be replaced with standard mean estimation.