Extrapolating noisy data: the challenging case with von Neumann entropy

Abstract:  Entanglement is a hallmark of quantum mechanics and describes the inherently non-local correlations between particles. To quantify this phenomenon, physicists often use the von Neumann entropy as a measure of entanglement. While theoretically well-understood, von Neumann entropy has proven notoriously difficult to measure experimentally. This talk presents a novel approach to extract von Neumann entropy from noisy, real-world data by extrapolating from the more accessible Rényi entropies. We first validate our technique using synthetic data, then put it to test by extracting the von Neumann entropy of a subsystem in a trapped-ion quantum simulator. We also discuss the adaptability of our protocol in broader scenarios, including the analytic continuation of Green’s functions away from the Matsubara domain in quantum Monte Carlo simulation, and measuring more sophisticated quantum information quantities such as mutual information.

Bio:Laimei Nie obtained her PhD from Stanford University and is currently an assistant professor at Purdue University. She is broadly interested in topics that are at the intersection of condensed matter physics and quantum information. Her current research interests include many-body quantum dynamics, such as quantum chaos, quantum thermalization, and localization, and how to measure entanglement in quantum simulators and solid-state systems.