Proper generalized decomposition for nonlinear convex problems in tensor Banach spaces

DOI10.1007/S00211-011-0437-5MaRDI QIDQ443852zbMATH OpenOpenAlexFDO

Authors Antonio Falcó, A. Nouy

Publication date 13 August 2012

Published in Numerische Mathematik (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1106.4424

zbMATH Keywords

convergencedecomposition methodsnonlinear Laplacianoptimization and variational techniquesprogressive and update proper generalized decompositionsingular value decompositiontensor Banach spacetensor products

Mathematics Subject Classification ID

Tensor products in functional analysis (46M05) Existence theories for problems in abstract spaces (49J27) Decomposition methods (49M27) Numerical optimization and variational techniques (65K10)

Abstract: Tensor-based methods are receiving a growing interest in scientific computing for the numerical solution of problems defined in high dimensional tensor product spaces. A family of methods called Proper Generalized Decompositions methods have been recently introduced for the a priori construction of tensor approximations of the solution of such problems. In this paper, we give a mathematical analysis of a family of progressive and updated Proper Generalized Decompositions for a particular class of problems associated with the minimization of a convex functional over a reflexive tensor Banach space.

Review text

The paper is well structured and very well written. It provides a concise overview about proper generalized decomposition methods (PGD) and allows a good understanding of the topic even for unexperienced readers. The presented results concerning progressive and updated PGDs are supported by numerous examples. After the introduction (Section 1), which includes a summary of already known results, the authors outline the goal of their research that is associated with the minimization of an elliptic and differentiable functional over some reflexive tensor Banach space. Therefore, the authors introduce the idea of progressive and updated PGDs. The progressive PGD consists of defining a sequence of approximates within some tensor subset where the next iterate is defined as an optimal correction of the previous iterate w.r.t. the underlying functional, whereas the updated PGD uses this optimal correction to span an affine subspace within which the next iterate is then defined as the minimizer of the functional. Therefore, the necessary terms and definitions for tensor Banach spaces are presented in Section 2 such as rank-one tensors, tensor Banach space etc. Furthermore, it is illustrated on several examples how such tensor Banach spaces embed into their original spaces and are weakly closed. Section 3 then provides classical results for the minimization of an functional over some reflexive Banach space such as the existence of a solution under certain assumptions, e.g. the functional is weakly sequential lower semi-continuous and the preimage being a bounded, weakly closed set in an reflexive Banach space or the uniqueness of the minimizer in case of a Fréchet differentiable and elliptic functional. Section 4 is the main part of the paper that provides a detailed description of the above mentioned progressive and updated PGD followed by the convergence analysis for Fréchet differentiable, elliptic functionals that are weakly sequentially continuous over a tensor Banach space. The last section (Section 5) before the conclusion includes two examples, a singular value decomposition in Lebesgue spaces and a nonlinear Laplacian.

Recommendations

• A proper generalized decomposition for the solution of elliptic problems in abstract form by using a functional Eckart-Young approach
• Numerical strategies for the Galerkin-proper generalized decomposition method
• Proper generalized decomposition for multiscale and~multiphysics problems
• The proper generalized decomposition for advanced numerical simulations. A primer
• On the computation of proper generalized decomposition modes of parametric elliptic problems

Cites work

• A generalized spectral decomposition technique to solve a class of linear stochastic partial differential equations
• A Multilinear Singular Value Decomposition
• A new family of solvers for some classes of multidimensional partial differential equations encountered in kinetic theory modeling of complex fluids
• A new scheme for the tensor representation
• A new tensor decomposition
• A priori model reduction through proper generalized decomposition for solving time-dependent partial differential equations
• A projection method to solve linear systems in tensor format
• A proper generalized decomposition for the solution of elliptic problems in abstract form by using a functional Eckart-Young approach
• Adaptive Optimization of Convex Functionals in Banach Spaces
• Differentiable functions on certain Banach spaces
• Functional analysis. An introduction to Banach space theory
• Generalized spectral decomposition method for solving stochastic finite element equations: invariant subspace problem and dedicated algorithms
• Greedy approximation
• scientific article; zbMATH DE number 1716480 (Why is no real title available?)
• scientific article; zbMATH DE number 3852340 (Why is no real title available?)
• scientific article; zbMATH DE number 3933858 (Why is no real title available?)
• scientific article; zbMATH DE number 108480 (Why is no real title available?)
• scientific article; zbMATH DE number 3525695 (Why is no real title available?)
• scientific article; zbMATH DE number 1250672 (Why is no real title available?)
• scientific article; zbMATH DE number 1077733 (Why is no real title available?)
• Mathematical methods in physics. Distributions, Hilbert space operators, variational methods, and applications in quantum physics
• On minimal subspaces in tensor representations
• On the convergence of a greedy rank-one update algorithm for a class of linear systems
• Orthogonal tensor decompositions
• Proper generalized decompositions and separated representations for the numerical solution of high dimensional stochastic problems
• Recent advances and new challenges in the use of the proper generalized decomposition for solving multidimensional models
• Recent developments in spectral stochastic methods for the numerical solution of stochastic partial differential equations
• Results and questions on a nonlinear approximation approach for solving high-dimensional partial differential equations
• Tensor Decompositions and Applications
• Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation Problem
• Tensor-structured Galerkin approximation of parametric and stochastic elliptic PDEs
• The LATIN multiscale computational method and the proper generalized decomposition
• Use of tensor formats in elliptic eigenvalue problems.
• Well-posedness of convex maximization problems on Stiefel manifolds and orthogonal tensor product approximations

Cited in

(41)

• On the existence of a progressive variational vademecum based on the proper generalized decomposition for a class of elliptic parameterized problems
• Solution of linear systems in high spatial dimensions
• A dynamical adaptive tensor method for the Vlasov-Poisson system
• On the convergence of alternating minimization methods in variational PGD
• Model reduction method using variable-separation for stochastic saddle point problems
• On the Dirac-Frenkel variational principle on tensor Banach spaces
• On minimal subspaces in tensor representations
• Tree-based tensor formats
• Numerical approximation of Poisson problems in long domains
• On the computation of proper generalized decomposition modes of parametric elliptic problems
• Editorial. Tensor numerical methods: actual theory and recent applications
• Iterative methods based on soft thresholding of hierarchical tensors
• Real time parameter identification and solution reconstruction from experimental data using the proper generalized decomposition
• A literature survey of low-rank tensor approximation techniques
• Nonincremental proper generalized decomposition solution of parametric uncoupled models defined in evolving domains
• Constrained optimization with low-rank tensors and applications to parametric problems with PDEs
• Greedy algorithms for high-dimensional non-symmetric linear problems
• A least-squares method for sparse low rank approximation of multivariate functions
• Tensor-based multiscale method for diffusion problems in quasi-periodic heterogeneous media
• Ideal minimal residual-based proper generalized decomposition for non-symmetric multi-field models -- application to transient elastodynamics in space-time domain
• A novel variable-separation method based on sparse and low rank representation for stochastic partial differential equations
• Tensor-based methods for numerical homogenization from high-resolution images
• Numerical tensor calculus
• A new algorithm of proper generalized decomposition for parametric symmetric elliptic problems
• Greedy algorithms for high-dimensional eigenvalue problems
• Towards a 2.5D geometric model in mold filling simulation
• Random field representations for stochastic elliptic boundary value problems and statistical inverse problems
• Tensor networks and hierarchical tensors for the solution of high-dimensional partial differential equations
• A Variable-Separation Method for Nonlinear Partial Differential Equations With Random Inputs
• Low-rank approximation of linear parabolic equations by space-time tensor Galerkin methods
• To be or not to be intrusive? The solution of parametric and stochastic equations -- proper generalized decomposition
• On direct sums and tensor products of non-linear programs
• Tensor numerical methods for multidimensional PDEs: theoretical analysis and initial applications
• Global convergence of rank-one PGD approximations by alternate minimization
• Low-rank tensor methods for partial differential equations
• Numerical Approximation of PGD Modes to Parameterized Elliptic Problems
• Structure and approximation properties of Laplacian-like matrices
• A proper generalized decomposition for the solution of elliptic problems in abstract form by using a functional Eckart-Young approach
• Universality for non-linear convex variational problems
• Ruminated tensor decomposition algorithm for solving inviscid Burgers' equation
• Adaptive near-optimal rank tensor approximation for high-dimensional operator equations

This page was built for publication: Proper generalized decomposition for nonlinear convex problems in tensor Banach spaces

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q443852)