Answers are grounded in this paper and cite their sources.
A wave function constructed as a product of a four-determinant function and a symmetric correlation function is employed in Monte Carlo computations of the ground-state energy of Li2 at Re = 5.05 Bohrs. Wave function parameters are determined by a fixed-sample minimization of deviations of the local energy. Although the variational Monte Carlo energy for this function lies, as expected, below that of a similar wave function constructed with a single determinant, the four-determinant function/correlation function wave function gives no improvement in quantum Monte Carlo energy. However, the unoptimized four-determinant function/correlation function wave function does yield an energy in excellent agreement with the estimated exact result. The poorer energy of the optimized function is caused by degradation of the nodal structure during parameter optimization.
No takes yet. Share an insight, caveat, or question.
Sun et al. (1992) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: