A hybrid Monte Carlo method
Jane L. Hutton
Abstract
Jane L. Hutton
Abstract
A method has been developed that combines the best features of deterministic and Monte Carlo techniques. Based on perturbation theory, this new method uses a deterministic solution of a simplified model along with a Monte Carlo calculation of the perturbation linking the simplified model to the actual complex geometry of the problem. If the perturbation is small, then the speed will be increased relative to a conventional Monte Carlo. A computer code MAX has been written, and the results of a range of lattice cell calculations are outlined that illustrate both the accuracy and efficiency of this technique.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A method has been developed that combines the best features of deterministic and Monte Carlo techniques. Based on perturbation theory, this new method uses a deterministic solution of a simplified model along with a Monte Carlo calculation of the perturbation linking the simplified model to the actual complex geometry of the problem. If the perturbation is small, then the speed will be increased relative to a conventional Monte Carlo. A computer code MAX has been written, and the results of a range of lattice cell calculations are outlined that illustrate both the accuracy and efficiency of this technique.
Key concepts: Monte Carlo method, Dynamic Monte Carlo method, Monte Carlo molecular modeling, Monte Carlo method in statistical physics, Hybrid Monte Carlo, Statistical physics, Monte Carlo integration, Quasi-Monte Carlo method