The sign problem and population dynamics in the full configuration interaction quantum Monte Carlo method

Area of Science:

• Computational Physics
• Quantum Chemistry
• Many-Body Physics

Background:

• Accurate calculation of ground-state energies for interacting fermion systems is computationally challenging.
• Traditional methods are limited by system size and the sign problem.
• The full configuration interaction quantum Monte Carlo (FCIQMC) method offers a potential solution.

Purpose of the Study:

• To investigate the nature and severity of the sign problem in FCIQMC.
• To understand how the sign problem's impact varies across different fermion systems.
• To elucidate the mechanism behind the convergence and population dynamics in FCIQMC simulations.

Main Methods:

• Application of the full configuration interaction quantum Monte Carlo (FCIQMC) method.
• Analysis of the sign problem's dependence on system characteristics.
• Examination of particle cancellation dynamics within the stochastic sampling.

Main Results:

• FCIQMC enables access to essentially exact ground-state energies for significantly larger fermion systems than previously possible.
• The severity of the sign problem is system-dependent.
• Cancellation between positive and negative sampled particles is crucial for convergence to the correct ground state.

Conclusions:

• FCIQMC is a powerful tool for studying large fermion systems without prior knowledge of nodal structure.
• Understanding the sign problem's behavior is key to optimizing FCIQMC simulations.
• The method's population dynamics are intrinsically linked to the cancellation mechanism ensuring accurate results.