Fast quantum measurement tomography with dimension-optimal error bounds

We present a two-step protocol for quantum measurement tomography that is light on classical co-processing cost and still achieves optimal sample complexity in the system dimension. Given measurement data from a known probe state ensemble, we first apply least-squares estimation to produce an unconstrained approximation of the POVM, and then project this estimate onto the set of valid quantum measurements. For a POVM with $L$ outcomes acting on a $d$-dimensional system, we show that the protocol requires $\mathcal{O}(d^3 L \ln(d)/\epsilon^2)$ samples to achieve error $\epsilon$ in worst-case distance, and $\mathcal{O}(d^2 L^2 \ln(dL)/\epsilon^2)$ samples in average-case distance. We further establish two almost matching sample complexity lower bounds of $\Omega(d^3/\epsilon^2)$ and $\Omega(d^2 L/\epsilon^2)$ for any non-adaptive, single-copy POVM tomography protocol. Hence, our projected least squares POVM tomography is sample-optimal in dimension $d$ up to logarithmic factors. Our method admits an analytic form when using global or local 2-designs as probe ensembles and enables rigorous non-asymptotic error guarantees. Finally, we also complement our findings with empirical performance studies carried out on a noisy superconducting quantum computer with flux-tunable transmon qubits.