Dynamics of Quantum Many-Body Systems from Monte Carlo simulations

Massimo Ostilli, Carlo Presilla A

Abstract. For many-body lattice quantum systems the existence of a Feynman-Kac–type formula based on Poisson stochastic processes has recently allowed to derive an exact Monte Carlo algorithm for the evolution operator at real or imaginary times. We extend this algorithm to the study of exact time-dependent correlation functions. The techniques generally employed in Monte Carlo simulations to control fluctuations, namely reconfigurations and importance sampling, are adapted to the present algorithm and their validity is rigorously proved. We complete the analysis by several examples for the hard-core boson Hubbard model and the Heisenberg model. PACS numbers: 02.50.-r, 02.70.SsExact Monte Carlo time dynamics in many-body lattice quantum systems 2 1.

Shiwei Zhang

David M. Ceperley

Since its first description fifty years ago, the Metropolis Monte Carlo method has been used in a variety of different ways for the simulation of continuum quantum many‐body systems. This paper will consider some of the generalizations of the Metropolis algorithm employed in quantum Monte Carlo: Variational Monte Carlo, dynamical methods for projector monte carlo (i.e. diffusion Monte Carlo with rejection), multilevel sampling in path integral Monte Carlo, the sampling of permutations, cluster methods for lattice models, the penalty method for coupled electron‐ionic systems and the Bayesian analysis of imaginary time correlation functions.

Wolfgang von der Linden

Yongdan Yang, Bing-Nan Lu, Ying Li

Quantum Monte Carlo and quantum simulation are both important tools for understanding quantum many-body systems. As a classical algorithm, quantum Monte Carlo suffers from the sign problem, preventing its applications to most fermion systems and real-time dynamics. In contrast, on a quantum computer, we can simulate time evolution of general systems with local interactions, yet such algorithms usually require the development of fault-tolerant quantum technologies. In this paper, we introduce a novel non-variational algorithm using quantum computing as a subroutine to accelerate quantum Monte Carlo. On a fault-tolerant quantum computer, the complexity of our algorithm is polynomial as the same as conventional quantum simulation algorithms. On a noisy quantum computer, our algorithm incorporating error mitigation eases the sign problem. Even when the circuit noise is significant, our algorithm can reduce the Monte Carlo variance by orders of magnitude.

Markus Holzmann