Sign Learning Kink-based (SiLK) Quantum Monte Carlo for molecular systems.

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Quantum Monte Carlo Methods

Background:

• The Quantum Monte Carlo (QMC) method is a powerful tool for calculating molecular energies.
• A significant challenge in QMC is the 'minus sign problem', which limits its accuracy and applicability.
• Accurate calculation of ground state energies is crucial for understanding molecular properties and reactions.

Purpose of the Study:

• To introduce and evaluate the Sign Learning Kink (SiLK) based Quantum Monte Carlo (QMC) method.
• To assess the SiLK method's ability to reduce or eliminate the QMC minus sign problem.
• To calculate ab initio ground state energies for H2O, N2, and F2 molecules with high accuracy.

Main Methods:

• Utilizing Feynman's path integral formulation of quantum mechanics.
• Implementing a two-stage approach: a learning stage to optimize Slater determinants and a conventional QMC simulation stage.
• Testing the SiLK method across different vector spaces.

Main Results:

• The SiLK method demonstrated accurate ab initio ground state energy calculations for H2O, N2, and F2.
• The method successfully reduced or eliminated the QMC minus sign problem.
• Results were validated by comparison with other quantum chemical methods and exact diagonalization.

Conclusions:

• The Sign Learning Kink (SiLK) based QMC method is a highly accurate computational chemistry technique.
• SiLK offers a viable solution to the long-standing QMC minus sign problem.
• This advancement has significant implications for future quantum mechanical simulations.