mVMC - Open-source software for many-variable variational Monte Carlo method

Takahiro Misawa, Satoshi Morita, Kazuyoshi Yoshimi, Mitsuaki Kawamura, Yuichi Motoyama, Kota Ido, Takahiro Ohgoe, Masatoshi Imada, Takeo Kato

Research output: Contribution to journalArticlepeer-review

Abstract

mVMC (many-variable Variational Monte Carlo) is an open-source software based on the variational Monte Carlo method applicable for a wide range of Hamiltonians for interacting fermion systems. In mVMC, we introduce more than ten thousands variational parameters and simultaneously optimize them by using the stochastic reconfiguration (SR) method. In this paper, we explain basics and user interfaces of mVMC. By using mVMC, users can perform the calculation by preparing only one input file of about ten lines for widely studied quantum lattice models, and can also perform it for general Hamiltonians by preparing several additional input files. We show the benchmark results of mVMC for the Hubbard model, the Heisenberg model, and the Kondo-lattice model. These benchmark results demonstrate that mVMC provides ground-state and low-energy-excited-state wave functions for interacting fermion systems with high accuracy.

Original languageEnglish
JournalUnknown Journal
Publication statusPublished - 2017 Nov 30

Keywords

  • 02.60.Dc Numerical linear algebra
  • 71.10.Fd Lattice fermion models

ASJC Scopus subject areas

  • General

Fingerprint Dive into the research topics of 'mVMC - Open-source software for many-variable variational Monte Carlo method'. Together they form a unique fingerprint.

Cite this