Unbiased reduced density matrices and electronic properties from full configuration interaction quantum Monte Carlo

Area of Science:

• Quantum chemistry
• Computational physics
• Electronic structure theory

Background:

• Calculating observable properties of electronic systems is crucial but difficult for projector quantum Monte Carlo methods.
• Pure (symmetric) expectation values are essential but computationally demanding.

Purpose of the Study:

• To develop an efficient method for sampling unbiased reduced density matrices in quantum Monte Carlo.
• To enable accurate computation of observable properties for electronic systems.

Main Methods:

• Utilizing an independent replica population of walkers within the full configuration interaction quantum Monte Carlo dynamic.
• Sampling alongside the original population to minimize computational overhead.

Main Results:

• The proposed method yields reduced density matrices free from systematic error.
• Rapid convergence to exact limits for expectation values and properties.
• A quasi-variational energy estimate accurately replaces projected estimators for multiconfigurational wavefunctions.

Conclusions:

• The novel approach provides accurate and unbiased reduced density matrices for quantum Monte Carlo simulations.
• The quasi-variational energy estimator offers a promising alternative for energy calculations.
• The method has potential for accurate extrapolation in larger quantum systems.