Abstract:Full Waveform Inversion (FWI) is a standard algorithm in seismic imaging. Its implementation requires the a priori choice of a number of "design parameters", such as the positions of sensors for the actual measurements and one (or more) regularisation weights. In this paper we describe a novel algorithm for determining these design parameters automatically from a set of training images, using a (supervised) bilevel learning approach. In our algorithm, the upper level objective function measures the quality of the reconstructions of the training images, where the reconstructions are obtained by solving the lower level optimisation problem -- in this case FWI. Our algorithm employs (variants of) the BFGS quasi-Newton method to perform the optimisation at each level, and thus requires the repeated solution of the forward problem -- here taken to be the Helmholtz equation. This paper focuses on the implementation of the algorithm. The novel contributions are: (i) an adjoint-state method for the efficient computation of the upper-level gradient; (ii) a complexity analysis for the bilevel algorithm, which counts the number of Helmholtz solves needed and shows this number is independent of the number of design parameters optimised; (iii) an effective preconditioning strategy for iteratively solving the linear systems required at each step of the bilevel algorithm; (iv) a smoothed extraction process for point values of the discretised wavefield, necessary for ensuring a smooth upper level objective function. The algorithm also uses an extension to the bilevel setting of classical frequency-continuation strategies, helping avoid convergence to spurious stationary points. The advantage of our algorithm is demonstrated on a problem derived from the standard Marmousi test problem.

Whiteness-based bilevel learning of regularization parameters in imaging

A general framework for whiteness-based parameters selection in variational models

Bilevel learning of regularization models and their discretization for image deblurring and super-resolution

Whiteness-based parameter selection for Poisson data in variational image processing

Residual whiteness principle for automatic parameter selection in $\ell_2$-$\ell_2$ image super-resolution problems

A Graph-based Iterative Algorithm for Inverse Problems with Local and Non-local Regularizations

Learning Sparsity-Promoting Regularizers Using Bilevel Optimization

Automatic nonstationary anisotropic Tikhonov regularization through bilevel optimization

Derivative-free stochastic bilevel optimization for inverse problems

Learning the optimal Tikhonov regularizer for inverse problems

Supervised Learning of Sparsity-Promoting Regularizers for Denoising

On Learning the Optimal Regularization Parameter in Inverse Problems

On the Asymptotical Regularization for Linear Inverse Problems in Presence of White Noise

Smoothing $\mathcal{L}^2$ gradients in iterative regularization

Enhancing deblurring through optimal controlled bilevel with nonlocal weighted model

Bilateral Spectrum Weighted Total Variation for Noisy-Image Super-Resolution and Image Denoising

Learning Weakly Convex Regularizers for Convergent Image-Reconstruction Algorithms

Unsupervised Training of Convex Regularizers using Maximum Likelihood Estimation

Basis Pursuit Denoise with Nonsmooth Constraints

Optimising seismic imaging design parameters via bilevel learning