Communication: An efficient stochastic algorithm for the perturbative density matrix renormalization group in large...

Area of Science:

• Quantum Chemistry
• Computational Physics

Background:

• The density matrix renormalization group (DMRG) method is powerful for strongly correlated systems.
• Large active spaces in molecular calculations pose significant computational challenges for deterministic DMRG algorithms.

Purpose of the Study:

• To introduce an efficient stochastic algorithm for the perturbative density matrix renormalization group (DMRG) method.
• To address the computational bottleneck in deterministic DMRG for large active spaces.

Main Methods:

• Developed a stochastic implementation of the perturbative DMRG method.
• Bypassed the computationally intensive step of solving first-order equations.

Main Results:

• Demonstrated the efficiency of the stochastic algorithm.
• Validated the accuracy of the new method on C2 and Cr2 molecular systems.
• The stochastic approach significantly reduces computational cost compared to deterministic methods.

Conclusions:

• The stochastic DMRG algorithm provides an efficient and accurate alternative for large active space calculations.
• This method offers a viable solution for tackling complex molecular systems previously intractable due to computational demands.