Efficient Classical Shadow Tomography through Many-body Localization Dynamics

Abstract: Classical shadow tomography serves as a potent tool for extracting numerous properties from quantum many-body systems with minimal measurements. Nevertheless, prevailing methods yielding optimal performance for few-body operators necessitate the application of random two-qubit gates, a task that can prove challenging on specific quantum simulators such as ultracold atomic gases. In this work, we introduce an alternative approach founded on the dynamics of many-body localization, a phenomenon extensively demonstrated in optical lattices. Through an exploration of the shadow norm -- both analytically, employing a phenomenological model, and numerically, utilizing the TEBD algorithm -- we demonstrate that our scheme achieves remarkable efficiency comparable to shallow circuits or measurement-induced criticality. This efficiency provides an exponential advantage over the Pauli measurement protocol for few-body measurements. Our findings are corroborated through direct numerical simulations encompassing the entire sampling and reconstruction processes. Consequently, our results present a compelling methodology for analyzing the output states of quantum simulators.