Multigrid renormalization

DOI10.1016/J.JCP.2018.06.065zbMATH Open1415.65273arXiv1802.07259OpenAlexW2796128092MaRDI QIDQ2000454FDOQ2000454

Pierre Moinier, Michael Lubasch, Dieter Jaksch

Publication date: 28 June 2019

Published in: Journal of Computational Physics (Search for Journal in Brave)

Abstract: We combine the multigrid (MG) method with state-of-the-art concepts from the variational formulation of the numerical renormalization group. The resulting MG renormalization (MGR) method is a natural generalization of the MG method for solving partial differential equations. When the solution on a grid of $N$ points is sought, our MGR method has a computational cost scaling as $mathcalO(log(N))$, as opposed to $mathcalO(N)$ for the best standard MG method. Therefore MGR can exponentially speed up standard MG computations. To illustrate our method, we develop a novel algorithm for the ground state computation of the nonlinear Schr"{o}dinger equation. Our algorithm acts variationally on tensor products and updates the tensors one after another by solving a local nonlinear optimization problem. We compare several different methods for the nonlinear tensor update and find that the Newton method is the most efficient as well as precise. The combination of MGR with our nonlinear ground state algorithm produces accurate results for the nonlinear Schr"{o}dinger equation on $N=10^{18}$ grid points in three spatial dimensions.

Full work available at URL: https://arxiv.org/abs/1802.07259

Recommendations

• The multigrid method for accelerated solution of the discretized Schrödinger equation
• An efficient multigrid method for ground state solution of Bose-Einstein condensates
• Multigrid and adaptive algorithm for solving the nonlinear Schrödinger equation
• scientific article; zbMATH DE number 18261
• scientific article; zbMATH DE number 769952

zbMATH Keywords

multigrid methodsdensity matrix renormalization groupmatrix product statesnumerical renormalization groupquantics tensor trainsvariational renormalization group methods

Mathematics Subject Classification ID

Multigrid methods; domain decomposition for boundary value problems involving PDEs (65N55) Laplace operator, Helmholtz equation (reduced wave equation), Poisson equation (35J05) NLS equations (nonlinear Schrödinger equations) (35Q55) Finite difference methods for boundary value problems involving PDEs (65N06)

Cites Work

• The tensor network theory library
• Tensor-Train Decomposition
• Simulating quantum systems on a quantum computer
• Multi-Level Adaptive Solutions to Boundary-Value Problems
• Title not available (Why is that?)
• Finitely correlated states on quantum spin chains
• Title not available (Why is that?)
• A practical introduction to tensor networks: Matrix product states and projected entangled pair states
• The density-matrix renormalization group
• Title not available (Why is that?)
• Renormalization and tensor product states in spin chains and lattices
• The density-matrix renormalization group in the age of matrix product states
• Approximation of $2^d\times2^d$ Matrices Using Tensor Decomposition
• Low-Rank Explicit QTT Representation of the Laplace Operator and Its Inverse
• Solution of Linear Systems and Matrix Inversion in the TT-Format
• \(O(d \log N)\)-quantics approximation of \(N\)-\(d\) tensors in high-dimensional numerical modeling
• Matrix product state representations
• Title not available (Why is that?)
• Constructive representation of functions in low-rank tensor formats
• Time Integration of Tensor Trains
• Entropy Scaling and Simulability by Matrix Product States
• Multigrid methods. Proceedings of the Conference Held at Köln-Porz, November 23-27, 1981
• Algebraically contractible topological tensor network states
• Unifying projected entangled pair state contractions

Cited In (2)

• Multiscale extensions for enhancing coarse grid computations
• Gauging tensor networks with belief propagation

Uses Software

• Tensor Network Theory

This page was built for publication: Multigrid renormalization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2000454)