An efficient perturbation theory of density matrix renormalization group
Density matrix renormalization group (DMRG) is one of the most powerful numerical methods available for many-body systems. In this work, we develop a perturbation theory of DMRG (PT-DMRG) to largely increase its accuracy in an extremely simple and efficient way. By using the canonical matrix product state (MPS) representation for the ground state of the considered system, a set of orthogonal basis functions $\left\lbrace | ψ_i \rangle \right\rbrace$ is introduced to describe the perturbations to the ground state obtained by the conventional DMRG. The Schmidt numbers of the MPS that are beyond the bond dimension cut-off are used to define such perturbation terms. The perturbed Hamiltonian is then defined as $\tilde{H}_{ij}= \langle ψ_i | \hat{H} | ψ_j \rangle$; its ground state permits to calculate physical observables with a considerably improved accuracy as compared to the original DMRG results. We benchmark the second-order perturbation theory with the help of one-dimensional Ising chain in a transverse field and the Heisenberg chain, where the precision of DMRG is shown to be improved $\rm O(10)$ times. Furthermore, for moderate length $L$ the errors of DMRG and PT-DMRG both scale linearly with $L^{-1}$. The linear relation between the dimension cut-off of DMRG and that of PT-DMRG with the same precision shows a considerable improvement of efficiency, especially for large dimension cut-off's. In thermodynamic limit we show that the errors of PT-DMRG scale with $\sqrt{L^{-1}}$. Our work suggests an effective way to define the tangent space of the ground state MPS, which may shed lights on the properties beyond the ground state. Such second-order PT-DMRG can be readily generalized to higher orders, as well as applied to the models in higher dimensions.