https://us02web.zoom.us/j/86848081459?pwd=ai8rWSt2ZWxFeDNtZ09PVFgzNU1jZz09

회의 ID: 868 4808 1459

암호: 076668

Simulating non-equilibrium quantum spin dynamics beyond one-dimension remains a significant challenge, despite advances in computational methods over the last two decades. We tackle this issue using a Monte Carlo method in which the time-evolution of  quantum observables is encoded by averages over a classical stochastic process. An intuitive approach is to sample around a "well-chosen" mean-field spin configuration. This works only for relatively short timescales, with entanglement growth generically requiring that these mean-fields decay over time. New insights are therefore required to reach later times. We demonstrate that in our stochastic representation one may encode mean-field contributions for every trajectory individually. This improves the efficiency of Monte Carlo sampling exponentially for quenches of the quantum Ising model in 2D and 3D. Connections to gauge freedom in phase space representations of quantum problems are also discussed, with emphasis placed on current and future research directions