Speaker

Mr. Mingcheng Yi  / The University of Tokyo

Title

Monte Carlo Tomography of von Neumann Entanglement Entropy: A Generative Reconstruction Approach Using Matrix Product Operators

Abstract

We propose a quantum Monte Carlo (QMC)–based framework for estimating the ground-state bipartite von Neumann entanglement entropy — a key quantity in characterizing quantum many-body entanglement, yet traditionally inaccessible to QMC methods that are typically limited to Rényi entropies. Our approach leverages reduced density matrices (RDMs) sampled via QMC and reconstructs the global state as a matrix product operator (MPO) using a tomographic, generative scheme. This reconstruction is guided by minimizing the Kullback–Leibler divergence between the marginal states of the target and the variational ansatz. By systematically comparing different protocols, we find that incorporating non-local RDMs combining nearest-neighbor pairs with a distant site enables the MPO to faithfully capture both quantum entanglement and long-range correlations simultaneously. We demonstrate accurate reproduction of two-point correlation functions and the central charge in a critical Ising system, suggesting broad applicability to other critical models. This approach holds promise for scalable entanglement estimation and may be extended to more complex quantum phases and/or higher-dimensions.

Profile of Speaker

• Sep. 2020 – Jun. 2024: Undergraduate student, Shanghai Jiao Tong University
• Oct. 2024 – Oct. 2025: Research Student, Institute for Solid State Physics, University of Tokyo
• From Oct. 2025: Master’s Student, the University of Tokyo

Join Zoom link

Meeting ID: 971 4015 7796
Passcode: 967548