An efficient duality-based approach for PDE-constrained sparse optimization

DOI10.1007/S10589-017-9951-4zbMATH Open1388.49037arXiv1708.09094OpenAlexW2963740496MaRDI QIDQ1744892FDOQ1744892

Authors: Xiaoliang Song, Bo Chen, Bo Yu 

Publication date: 20 April 2018

Published in: Computational Optimization and Applications (Search for Journal in Brave)

Abstract: In this paper, elliptic optimal control problems involving the $L^1$-control cost ($L^1$-EOCP) is considered. To numerically discretize $L^1$-EOCP, the standard piecewise linear finite element is employed. However, different from the finite dimensional $l^1$-regularization optimization, the resulting discrete $L^1$-norm does not have a decoupled form. A common approach to overcome this difficulty is employing a nodal quadrature formula to approximately discretize the $L^1$-norm. It is clear that this technique will incur an additional error. To avoid the additional error, solving $L^1$-EOCP via its dual, which can be reformulated as a multi-block unconstrained convex composite minimization problem, is considered. Motivated by the success of the accelerated block coordinate descent (ABCD) method for solving large scale convex minimization problems in finite dimensional space, we consider extending this method to $L^1$-EOCP. Hence, an efficient inexact ABCD method is introduced for solving $L^1$-EOCP. The design of this method combines an inexact 2-block majorized ABCD and the recent advances in the inexact symmetric Gauss-Seidel (sGS) technique for solving a multi-block convex composite quadratic programming whose objective contains a nonsmooth term involving only the first block. The proposed algorithm (called sGS-imABCD) is illustrated at two numerical examples. Numerical results not only confirm the finite element error estimates, but also show that our proposed algorithm is more efficient than (a) the ihADMM (inexact heterogeneous alternating direction method of multipliers), (b) the APG (accelerated proximal gradient) method.

Full work available at URL: https://arxiv.org/abs/1708.09094

Recommendations

• Error estimates for sparse optimal control problems by piecewise linear finite element approximation
• An FE-inexact heterogeneous ADMM for elliptic optimal control problems with \(L^1\)-control cost
• PDE-constrained optimization: optimal control with \(L_1\)-regularization, state and control box constraints
• Preconditioning PDE-constrained optimization with \(L^1\)-sparsity and control constraints
• Convergence and regularization results for optimal control problems with sparsity functional

zbMATH Keywords

sparsityoptimal controlfinite elementduality approachaccelerated block coordinate descent

Mathematics Subject Classification ID

Discrete approximations in optimal control (49M25) Numerical methods involving duality (49M29) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Distributed algorithms (68W15) Large-scale systems (93A15) Linear optimal control problems (49N05)

Cites Work

• A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
• Hankel matrix rank minimization with applications to system identification and realization
• An efficient inexact symmetric Gauss-Seidel based majorized ADMM for high-dimensional convex composite conic programming
• A Schur complement based semi-proximal ADMM for convex quadratic conic programming and extensions
• A variational discretization concept in control constrained optimization: The linear-quadratic case
• An accelerated proximal gradient algorithm for nuclear norm regularized linear least squares problems
• Optimization with PDE Constraints
• On the approximation of the solution of an optimal control problem governed by an elliptic equation
• Title not available (Why is that?)
• An inexact accelerated proximal gradient method for large scale linearly constrained convex SDP
• Approximation of a class of optimal control problems with order of convergence estimates
• Optimal solvers for PDE-constrained optimization
• Preconditioned Conjugate Gradient Method for Optimal Control Problems with Control and State Constraints
• Error estimates for linear-quadratic control problems with control constraints
• Primal-Dual Strategy for Constrained Optimal Control Problems
• Superconvergence Properties of Optimal Control Problems
• Preconditioned MHSS iteration methods for a class of block two-by-two linear systems with applications to distributed control problems
• Iterative thresholding for sparse approximations
• Using piecewise linear functions in the numerical approximation of semilinear elliptic control problems
• Convergence and regularization results for optimal control problems with sparsity functional
• Optimality conditions and error analysis of semilinear elliptic control problems with \(L^1\) cost functional
• A mesh-independence result for semismooth Newton methods.
• Finite elements and fast iterative solvers. With applications in incompressible fluid dynamics
• Chebyshev semi-iteration in preconditioning for problems including the mass matrix
• A duality-based approach to elliptic control problems in non-reflexive Banach spaces
• Semismooth Newton Methods for Operator Equations in Function Spaces
• Generalized Hessian matrix and second-order optimality conditions for problems with \(C^{1,1}\) data
• Approximation of sparse controls in semilinear equations by piecewise linear functions
• Elliptic optimal control problems with \(L^1\)-control cost and applications for the placement of control devices
• Title not available (Why is that?)
• Realistic Eigenvalue Bounds for the Galerkin Mass Matrix
• An efficient inexact ABCD method for least squares semidefinite programming
• Approximation of elliptic control problems in measure spaces with sparse solutions
• Preconditioning PDE-constrained optimization with \(L^1\)-sparsity and control constraints

Cited In (6)

• A global convergent semi-smooth Newton method for semi-linear elliptic optimal control problem
• A diagonal finite element-projection-proximal gradient algorithm for elliptic optimal control problem
• A Zeroth-Order Proximal Stochastic Gradient Method for Weakly Convex Stochastic Optimization
• Smoothing Analysis of Two Robust Multigrid Methods for Elliptic Optimal Control Problems
• Inexact primal-dual active set iteration for optimal distribution control of stationary heat or cold source
• Preconditioning for PDE-constrained optimization with total variation regularization

Uses Software

• iFEM
• QSDPNAL

This page was built for publication: An efficient duality-based approach for PDE-constrained sparse optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1744892)