Abstract

We discuss that flat-histogram techniques can be appropriately applied in the sampling of quantum Monte Carlo simulation in order to improve the statistical quality of the results at long imaginary time or low excitation energy. Typical imaginary-time correlation functions calculated in quantum Monte Carlo are subject to exponentially growing errors as the range of imaginary time grows and this smears the information on the low energy excitations. We show that we can extract the low energy physics by modifying the Monte Carlo sampling technique to one in which configurations which contribute to making the histogram of certain quantities flat are promoted. We apply the diagrammatic Monte Carlo (diag-MC) method to the motion of a single hole in the t-J model and we show that the implementation of flat-histogram techniques allows us to calculate the Green's function in a wide range of imaginary-time. In addition, we show that applying the flat-histogram technique alleviates the “sign”-problem associated with the simulation of the single-hole Green's function at long imaginary time.

Details

Title
Flat-histogram methods in quantum Monte Carlo simulations: Application to the t-J model
Author
Diamantis, Nikolaos G 1 ; Manousakis, Efstratios 2 

 Department of Physics, University of Athens, Panepistimioupolis, Zografos, 157 84 Athens, Greece 
 Department of Physics, University of Athens, Panepistimioupolis, Zografos, 157 84 Athens, Greece; Department of Physics and National High Magnetic Field Laboratory, Florida State University, Tallahassee, FL 32306-4350, USA 
Publication year
2016
Publication date
Sep 2016
Publisher
IOP Publishing
ISSN
17426588
e-ISSN
17426596
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2575213365
Copyright
© 2016. This work is published under http://creativecommons.org/licenses/by/3.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.