Self-optimizing surplus function quantum Monte Carlo approach

This paper proposes a novel surplus function quantum Monte Carlo algorithm, which is a self-optimizing and self-improving procedure. In contrast to previous procedure, the trial function is optimized synchronistically in the surplus function method, but not before the variational Monte Carlo computation. In order to optimize the trial function, an improved steepest descent technique is used, with the step size automatically adjusted to obtain a procedure which converges superlinearly. The author also uses a novel correlation function, which has both the correct electron-electron and electron-nucleus cusp conditions. The method is employed to calculate the ground state energies of H-2, LiH, Li-2, H2O and the energy values of X B-3(1) state, 1 (1)A(1) state, and 2 (1)A(1) state of CH2.