Self-Learning Monte Carlo Method in Fermion Systems
Strongly Correlated Electrons
2017-06-14 v1 Disordered Systems and Neural Networks
Statistical Mechanics
Abstract
We develop the self-learning Monte Carlo (SLMC) method, a general-purpose numerical method recently introduced to simulate many-body systems, for studying interacting fermion systems. Our method uses a highly-efficient update algorithm, which we design and dub "cumulative update", to generate new candidate configurations in the Markov chain based on a self-learned bosonic effective model. From general analysis and numerical study of the double exchange model as an example, we find the SLMC with cumulative update drastically reduces the computational cost of the simulation, while remaining statistically exact. Remarkably, its computational complexity is far less than the conventional algorithm with local updates.
Cite
@article{arxiv.1611.09364,
title = {Self-Learning Monte Carlo Method in Fermion Systems},
author = {Junwei Liu and Huitao Shen and Yang Qi and Zi Yang Meng and Liang Fu},
journal= {arXiv preprint arXiv:1611.09364},
year = {2017}
}