7 March 1994 in Physical Review Letters
We present a powerful method for calculating the thermodynamic properties of infinite-dimensional Hubbard-type models using an exact diagonalization of an Anderson model with a finite number of sites. The resolution obtained for Green’s functions is far superior to that of quantum Monte Carlo calculations. We apply the method to the half-filled Hubbard model for a discussion of the metal-insulator transition, and to the two-band Hubbard model where we find direct evidence for the existence of a superconducting instability at low temperatures.
1 February 1992 in Physical Review B
We study a model of strongly interacting lattice bosons with a Gutzwiller-type wave function that contains only on-site correlations. The variational energy and the condensate fraction associated with the variational wave function are exactly evaluated for both finite and infinite systems and compared with exact quantum Monte Carlo results in two dimensions. This ansatz for the wave function gives the correct qualitative picture of the phase diagram of this system; at commensurate densities, this system enters a Mott-insulator phase for large values of the interaction.
1 September 2013 in Canadian Journal of Chemistry
Defining accurate and compact trial wavefunctions leading to small statistical and fixed-node errors in quantum Monte Carlo (QMC) calculations is still a challenging problem. Here we propose to make use of selected configuration interaction (CI) expansions obtained by selecting the most important determinants through a perturbative criterion. A major advantage with respect to truncated CASSCF wavefunctions or CI expansions limited to a maximum number of excitations (e.g, CISD) is that much smaller expansions can be considered (many unessential determinants are avoided), an important practical point for efficient QMC calculations. The most important determinants entering first during the selection process (hierarchical construction) the main features of the nodal structure of the wavefunction can be expected to be obtained with a moderate number of determinants. Thanks to this property, the delicate problem of optimizing in a Monte Carlo framework the numerous linear and (or) nonlinear parameters of the determinantal part of the trial wavefunction could be avoided. As a first numerical example, the calculation of the ground-state energy of the oxygen atom is presented. The best DMC value reported so far is obtained.
6 December 1999 in Physical Review Letters
We present a general approach to greatly increase at little cost the efficiency of Monte Carlo algorithms. To each observable to be computed we associate a renormalized observable (improved estimator) having the same average but a different variance. By writing down the zero-variance condition a fundamental equation determining the optimal choice for the renormalized observable is derived (zero-variance principle for each observable separately). We show, with several examples including classical and quantum Monte Carlo calculations, that the method can be very powerful.
17 July 2017 in The Journal of Chemical Physics
A hybrid stochastic-deterministic approach for computing the second-order perturbative contribution E(2) within multireference perturbation theory (MRPT) is presented. The idea at the heart of our hybrid scheme—based on a reformulation of E(2) as a sum of elementary contributions associated with each determinant of the MR wave function—is to split E(2) into a stochastic and a deterministic part. During the simulation, the stochastic part is gradually reduced by dynamically increasing the deterministic part until one reaches the desired accuracy. In sharp contrast with a purely stochastic Monte Carlo scheme where the error decreases indefinitely as t−1/2 (where t is the computational time), the statistical error in our hybrid algorithm displays a polynomial decay with n = 3–4 in the examples considered here. If desired, the calculation can be carried on until the stochastic part entirely vanishes. In that case, the exact result is obtained with no error bar and no noticeable computational overhead compared to the fully deterministic calculation. The method is illustrated on the and molecules. Even for the largest case corresponding to the molecule treated with the cc-pVQZ basis set, very accurate results are obtained for E(2) for an active space of (28e, 176o) and a MR wave function including up to determinants.
8 August 2018 in The Journal of Chemical Physics
Selected configuration interaction (sCI) methods including second-order perturbative corrections provide near full CI (FCI) quality energies with only a small fraction of the determinants of the FCI space. Here, we introduce both a state-specific and a multi-state sCI method based on the configuration interaction using a perturbative selection made iteratively (CIPSI) algorithm. The present method revises the reference (internal) space under the effect of its interaction with the outer space via the construction of an effective Hamiltonian, following the shifted-Bk philosophy of Davidson and co-workers. In particular, the multi-state algorithm removes the storage bottleneck of the effective Hamiltonian via a low-rank factorization of the dressing matrix. Illustrative examples are reported for the state-specific and multi-state versions.