2016
DOI: 10.1063/1.4954726
|View full text |Cite
|
Sign up to set email alerts
|

Quantum Monte Carlo with variable spins

Abstract: We investigate the inclusion of variable spins in electronic structure quantum Monte Carlo, with a focus on diffusion Monte Carlo with Hamiltonians that include spin-orbit interactions. Following our previous introduction of fixed-phase spin-orbit diffusion Monte Carlo (FPSODMC), we thoroughly discuss the details of the method and elaborate upon its technicalities. We present a proof for an upper-bound property for complex nonlocal operators, which allows for the implementation of T-moves to ensure the variati… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...

Citation Types

3
43
0

Year Published

2016
2016
2023
2023

Publication Types

Select...
6
1

Relationship

3
4

Authors

Journals

citations
Cited by 22 publications
(46 citation statements)
references
References 49 publications
3
43
0
Order By: Relevance
“…The resulting energy is not necessarily variational, however, the variational property can be recovered with an appropriate modification [13] of the T-moves algorithm [14]. Continuous spin and its updates.…”
mentioning
confidence: 99%
See 2 more Smart Citations
“…The resulting energy is not necessarily variational, however, the variational property can be recovered with an appropriate modification [13] of the T-moves algorithm [14]. Continuous spin and its updates.…”
mentioning
confidence: 99%
“…The Jastrow factor includes electron-ion, electron-electron and, possibly, higher order terms. Since U (R) depends only on the spatial coordinates, the spin integrations in the nonlocal operator can be done explicitly and the rest is similar to the treatment of nonlocality in static spin calculations [11,13]. The short-time approximation for the importance sampled propagator [3,4] is a product of the dynamical and reweighting factors G(R, S; R , S ) = G dyn e −∆t(E loc +E loc −2E T )/2 , where E loc (R, S) = [HΨ T ]/Ψ T .…”
mentioning
confidence: 99%
See 1 more Smart Citation
“…Thanks to well established advances 9,10 in this field, it is possible nowadays to compute the total energy of a given correlated ansatz and to optimize several variational parameters with a computational effort scaling at most with the fourth power of the number of electrons.A good variational ansatz allows a good description of the ground state by energy optimization. Moreover an even better characterization can be obtain by applying the so called diffusion Monte Carlo (DMC) method with the Fixed Node approximation (FNA) 11,12 . Within this projection method it is possible to obtain the lowest energy state constrained to have the same signs of a chosen trial WF, in the configuration space where electron positions and spins are given.…”
mentioning
confidence: 99%
“…A good variational ansatz allows a good description of the ground state by energy optimization. Moreover an even better characterization can be obtain by applying the so called diffusion Monte Carlo (DMC) method with the Fixed Node approximation (FNA) 11,12 . Within this projection method it is possible to obtain the lowest energy state constrained to have the same signs of a chosen trial WF, in the configuration space where electron positions and spins are given.…”
mentioning
confidence: 99%