Tensor network state methods and quantum information theory for strongly correlated molecular systems

TL;DR

This paper reviews recent tensor network state methods and their application to strongly correlated molecular systems, highlighting advances in optimization, parallelization, and symmetry handling for improved computational efficiency.

Contribution

It introduces new tensor network techniques and computational strategies specifically tailored for strongly correlated molecular systems, enhancing accuracy and scalability.

Findings

Global fermionic mode optimization improves wave function parametrization.

Hybrid CPU-multiGPU parallelization accelerates computations.

Efficient treatment of non-Abelian symmetries enhances high-performance calculations.

Abstract

A brief pedagogical overview of recent advances in tensor network state methods are presented that have the potential to broaden their scope of application radically for strongly correlated molecular systems. These include global fermionic mode optimization, i.e., a general approach to find an optimal matrix product state (MPS) parametrization of a quantum many-body wave function with the minimum number of parameters for a given error margin, the restricted active space DMRG-RAS-X method, multi-orbital correlations and entanglement, developments on hybrid CPU-multiGPU parallelization, and an efficient treatment of non-Abelian symmetries on high-performance computing (HPC) infrastructures. Scaling analysis on NVIDIA DGX-A100 platform is also presented.