A sparse control approach to optimal sensor placement in PDE-constrained parameter estimation problems

We present a systematic approach to the optimal placement of finitely many sensors in order to infer a finite-dimensional parameter from point evaluations of the solution of an associated parameter-dependent elliptic PDE. The quality of the corresponding least squares estimator is quantified by properties of the asymptotic covariance matrix depending on the distribution of the measurement sensors. We formulate a design problem where we minimize functionals related to the size of the corresponding confidence regions with respect to the position and number of pointwise measurements. The measurement setup is modeled by a positive Borel measure on the spatial experimental domain resulting in a convex optimization problem. For the algorithmic solution a class of accelerated conditional gradient methods in measure space is derived, which exploits the structural properties of the design problem to ensure convergence towards sparse solutions. Convergence properties are presented and the presented results are illustrated by numerical experiments.

[1]  J. Dunn.  Rates of convergence for conditional gradient algorithms near singular and nonsingular extremals , 1978 .

[2]  R. Cooke.  Real and Complex Analysis , 2011 .

[3]  Georg Stadler,et al.  Elliptic optimal control problems with L1-control cost and applications for the placement of control devices , 2009, Comput. Optim. Appl..

[4]  E. Haber,et al.  Numerical methods for experimental design of large-scale linear ill-posed inverse problems , 2008 .

[5]  Steven E. Rigdon,et al.  Model-Oriented Design of Experiments , 1997, Technometrics.

[6]  C. Atwood.  Sequences Converging to $D$-Optimal Designs of Experiments , 1973 .

[7]  Yaming Yu,et al.  D-optimal designs via a cocktail algorithm , 2009, Stat. Comput..

[8]  Luc Pronzato,et al.  Removing non-optimal support points in D-optimum design algorithms , 2003 .

[9]  G. Burton.  Sobolev Spaces , 2013 .

[10]  J. Kiefer.  General Equivalence Theory for Optimum Designs (Approximate Theory) , 1974 .

[11]  Michael Hintermüller,et al.  The Length of the Primal-Dual Path in Moreau-Yosida-Based Path-Following Methods for State Constrained Optimal Control , 2014, SIAM J. Optim..

[12]  K. Bredies,et al.  Inverse problems in spaces of measures , 2013 .

[13]  Michael Jackson,et al.  Optimal Design of Experiments , 1994 .

[14]  Nicolas Courty,et al.  Generalized conditional gradient: analysis of convergence and applications , 2015, ArXiv.

[15]  M. Pachter,et al.  Optimal control of partial differential equations , 1980 .

[16]  Heinz Bauer,et al.  Maß- und Integrationstheorie , 1992 .

[17]  J. Kiefer,et al.  Optimum Designs in Regression Problems , 1959 .

[18]  Martin Jaggi,et al.  Revisiting Frank-Wolfe: Projection-Free Sparse Convex Optimization , 2013, ICML.

[19]  E. Steinitz.  Bedingt konvergente Reihen und konvexe Systeme. , 1913 .

[20]  Roland Herzog,et al.  Approximation of sparse controls in semilinear equations by piecewise linear functions , 2012, Numerische Mathematik.

[21]  J. Graver,et al.  Graduate studies in mathematics , 1993 .

[22]  Philip Wolfe,et al.  An algorithm for quadratic programming , 1956 .

[23]  Karl Kunisch,et al.  Path-following Methods for a Class of Constrained Minimization Problems in Function Space , 2006, SIAM J. Optim..

[24]  Konstantin Pieper,et al.  Finite element discretization and efficient numerical solution of elliptic and parabolic sparse control problems , 2015 .

[25]  H. Wynn.  The Sequential Generation of $D$-Optimum Experimental Designs , 1970 .

[26]  Andrej Pázman,et al.  Foundations of Optimum Experimental Design , 1986 .

[27]  Roland Herzog,et al.  Sequentially optimal sensor placement in thermoelastic models for real time applications , 2015 .

[28]  W. Näther.  Optimum experimental designs , 1994 .

[29]  Stefan Wendl,et al.  Optimal Control of Partial Differential Equations , 2021, Applied Mathematical Sciences.

[30]  Gabriel Peyré,et al.  Exact Support Recovery for Sparse Spikes Deconvolution , 2013, Foundations of Computational Mathematics.

[31]  Anthony C. Atkinson,et al.  Optimum Experimental Designs, with SAS , 2007 .

[32]  Benjamin Pfaff,et al.  Perturbation Analysis Of Optimization Problems , 2016 .

[33]  H. Banks,et al.  Experimental design for vector output systems , 2014, Inverse problems in science and engineering.

[34]  G. Milliken.  Nonlinear Regression Analysis and Its Applications , 1990 .

[35]  E. Beale,et al.  Confidence Regions in Non‐Linear Estimation , 1960 .

[36]  Ernst Steinitz.  Bedingt konvergente Reihen und konvexe Systeme. (Fortsetzung). , 1914 .

[37]  K. Kunisch,et al.  A duality-based approach to elliptic control problems in non-reflexive Banach spaces , 2011 .

[38]  Boris Vexler,et al.  Adaptive Finite Element Methods for Parameter Identification Problems , 2013 .

[39]  Karl Kunisch,et al.  Feasible and Noninterior Path-Following in Constrained Minimization with Low Multiplier Regularity , 2006, SIAM J. Control. Optim..

[40]  Georg Stadler,et al.  A-Optimal Design of Experiments for Infinite-Dimensional Bayesian Linear Inverse Problems with Regularized ℓ0-Sparsification , 2013, SIAM J. Sci. Comput..

[41]  D. Ucinski.  Optimal measurement methods for distributed parameter system identification , 2004 .

[42]  Albert Tarantola,et al.  Inverse problem theory - and methods for model parameter estimation , 2004 .

[43]  Roland Becker,et al.  Parameter identification for chemical models in combustion problems , 2005 .

[44]  H. Brezis.  Functional Analysis, Sobolev Spaces and Partial Differential Equations , 2010 .

[45]  Eldad Haber,et al.  Experimental Design for Biological Systems , 2012, SIAM J. Control. Optim..

[46]  Michael Ulbrich,et al.  A Semismooth Newton Method with Multidimensional Filter Globalization for l1-Optimization , 2014, SIAM J. Optim..

[47]  Michael Hinze,et al.  A Variational Discretization Concept in Control Constrained Optimization: The Linear-Quadratic Case , 2005, Comput. Optim. Appl..

[48]  Hans Bock,et al.  Numerical methods for optimum experimental design in DAE systems , 2000 .

[49]  R. C. St. John,et al.  D-Optimality for Regression Designs: A Review , 1975 .

[50]  Konstantin Pieper,et al.  Inverse point source location with the Helmholtz equation on a bounded domain , 2018, Computational Optimization and Applications.

[51]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[52]  Hans Bock,et al.  A sequential approach for nonlinear optimum experimental design in DAE systems , 1999 .

[53]  Benjamin Recht,et al.  The alternating descent conditional gradient method for sparse inverse problems , 2015, 2015 IEEE 6th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP).

[54]  J. Dunn.  Convergence Rates for Conditional Gradient Sequences Generated by Implicit Step Length Rules , 1980 .

[55]  Dirk A. Lorenz,et al.  A generalized conditional gradient method and its connection to an iterative shrinkage method , 2009, Comput. Optim. Appl..

[56]  Valerii V. Fedorov,et al.  Optimal Design for Nonlinear Response Models , 2013 .

[57]  Karl Kunisch,et al.  Approximation of Elliptic Control Problems in Measure Spaces with Sparse Solutions , 2012, SIAM J. Control. Optim..

[58]  Douglas M. Bates,et al.  Nonlinear Regression Analysis and Its Applications , 1988 .

[59]  W. J. Studden,et al.  Theory Of Optimal Experiments , 1972 .

[60]  F. Tröltzsch.  Optimal Control of Partial Differential Equations: Theory, Methods and Applications , 2010 .

[61]  Michael Ulbrich,et al.  Semismooth Newton Methods for Operator Equations in Function Spaces , 2002, SIAM J. Optim..

[62]  H. Banks,et al.  Experimental design and inverse problems in plant biological modeling , 2012 .