Abstract:We address the problem of optimal experimental design (OED) for Bayesian nonlinear inverse problems governed by PDEs. The goal is to find a placement of sensors, at which experimental data are collected, so as to minimize the uncertainty in the inferred parameter field. We formulate the OED objective function by generalizing the classical A-optimal experimental design criterion using the expected value of the trace of the posterior covariance. We seek a method that solves the OED problem at a cost (measured in the number of forward PDE solves) that is independent of both the parameter and sensor dimensions. To facilitate this, we construct a Gaussian approximation to the posterior at the maximum a posteriori probability (MAP) point, and use the resulting covariance operator to define the OED objective function. We use randomized trace estimation to compute the trace of this (implicitly defined) covariance operator. The resulting OED problem includes as constraints the PDEs characterizing the MAP point, and the PDEs describing the action of the covariance operator to vectors. The sparsity of the sensor configurations is controlled using sparsifying penalty functions. We elaborate our OED method for the problem of determining the sensor placement to best infer the coefficient of an elliptic PDE. Adjoint methods are used to compute the gradient of the PDE-constrained OED objective function. We provide numerical results for inference of the permeability field in a porous medium flow problem, and demonstrate that the number of PDE solves required for the evaluation of the OED objective function and its gradient is essentially independent of both the parameter and sensor dimensions. The number of quasi-Newton iterations for computing an OED also exhibits the same dimension invariance properties.

On the Construction of an Optimal Network of Observation Points when Solving Inverse Linear Problems of Gravimetry and Magnetometry

Solving Some Inverse Problems of Gravimetry and Magnetometry Using an Algorithm That Improves Matrix Conditioning

Solution of the Interpretation Tomography Problem in Geophysics under the Linear Integral Representation Method and Theories of Discrete Gravity and the Magnetic Potential

Geometric methods for optimal sensor design

Lavrent'ev-Type Equations and Systems in the Inverse Problem of Reconstructing Viscoelastic Medium Memory

On the construction of families of optimal recovery methods for linear operators

On Constructing of Magnetic and Gravity Images of Mercury from Satellite Data

On the Simultaneous Determination of the Distribution Density of Sources Equivalent in the External Field and the Spectrum of the Useful Signal

A Fast and Scalable Method for A-Optimal Design of Experiments for Infinite-dimensional Bayesian Nonlinear Inverse Problems

Optimal Experimental Design for Inverse Problems in the Presence of Observation Correlations

Ionospheric radio tomography based on the GPS/GLONASS navigation systems

Global convergence of iterative solvers for problems of nonlinear magnetostatics

Satellite measurement errors and their impact on determination of point coordinates

M. M. Lavrentiev-type systems and reconstructing parameters of viscoelastic media

On the Uniqueness of the Solution to the Inverse Problem of Determining the Diffusion Coefficient of the Magnetic Field Necessary for Constructing Analytical Models of the Magnetic Field of Mercury

On the Features of Numerical Solution of Coefficient Inverse Problems for Nonlinear Equations of the Reaction–Diffusion–Advection Type with Data of Various Types

On Exact Solutions of a Multidimensional System of Elliptic Equations with Power-Law Nonlinearities

Optimization Methods for Inverse Problems

On an Iterative Method of Solving Direct and Inverse Problems for Parabolic Equations

Distributed Least-Squares Iterative Methods in Networks: A Survey.

Robust A-Optimal Experimental Design for Bayesian Inverse Problems