Hierarchical off-diagonal low-rank approximation of Hessians in inverse problems, with application to ice sheet model initialization
From MaRDI portal
Publication:6166000
DOI10.1088/1361-6420/ACD719arXiv2301.03644MaRDI QIDQ6166000FDOQ6166000
Authors: Tucker Hartland, Georg Stadler, Mauro Perego, Kim Liegeois, Noemi Petra 
Publication date: 6 July 2023
Published in: Inverse Problems (Search for Journal in Brave)
Abstract: Obtaining lightweight and accurate approximations of Hessian applies in inverse problems governed by partial differential equations (PDEs) is an essential task to make both deterministic and Bayesian statistical large-scale inverse problems computationally tractable. The computational complexity of dense linear algebraic routines such as that needed for sampling from Gaussian proposal distributions and Newton solves by direct linear methods, can be reduced to log-linear complexity by utilizing hierarchical off-diagonal low-rank (HODLR) matrix approximations. In this work, we show that a class of Hessians that arise from inverse problems governed by PDEs are well approximated by the HODLR matrix format. In particular, we study inverse problems governed by PDEs that model the instantaneous viscous flow of ice sheets. In these problems, we seek a spatially distributed basal sliding parameter field such that the flow predicted by the ice sheet model is consistent with ice sheet surface velocity observations. We demonstrate the use of HODLR approximation by efficiently generating Hessian approximations that allow fast generation of samples from a Gaussianized posterior proposal distribution. Computational studies are performed which illustrate ice sheet problem regimes for which the Gauss-Newton data-misfit Hessian is more efficiently approximated by the HODLR matrix format than the low-rank (LR) format. We then demonstrate that HODLR approximations can be favorable, when compared to global low-rank approximations, for large-scale problems by studying the data-misfit Hessian associated to inverse problems governed by the Stokes flow model on the Humboldt glacier and Greenland ice sheets.
Full work available at URL: https://arxiv.org/abs/2301.03644
Numerical methods for partial differential equations, boundary value problems (65Nxx) Numerical linear algebra (65Fxx) Probabilistic methods, stochastic differential equations (65Cxx)
Cites Work
- Title not available (Why is that?)
- An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach
- A sparse matrix arithmetic based on \({\mathfrak H}\)-matrices. I: Introduction to \({\mathfrak H}\)-matrices
- Mitigating the influence of the boundary on PDE-based covariance operators
- Data-sparse approximation by adaptive \({\mathcal H}^2\)-matrices
- An \(\mathcal O(N\log N)\) fast direct solver for partial hierarchically semi-separable matrices. With application to radial basis function interpolation
- Monte Carlo sampling methods using Markov chains and their applications
- Compressing Rank-Structured Matrices via Randomized Sampling
- Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
- Scalable Gaussian Process Computations Using Hierarchical Matrices
- Title not available (Why is that?)
- Inverse problems: a Bayesian perspective
- Nonstationary inverse problems and state estimation
- Title not available (Why is that?)
- Title not available (Why is that?)
- Inverse Problem Theory and Methods for Model Parameter Estimation
- A computational framework for infinite-dimensional Bayesian inverse problems. I: The linearized case, with application to global seismic inversion
- Fast algorithms for Bayesian uncertainty quantification in large-scale linear inverse problems based on low-rank partial Hessian approximations
- A computational framework for infinite-dimensional Bayesian inverse problems. II: stochastic Newton MCMC with application to ice sheet flow inverse problems
- A fast randomized algorithm for computing a hierarchically semiseparable representation of a matrix
- Fast construction of hierarchical matrix representation from matrix-vector multiplication
- On a generalization of the preconditioned Crank-Nicolson metropolis algorithm
- Algorithms for Kullback-Leibler approximation of probability measures in infinite dimensions
- Optimal low-rank approximations of Bayesian linear inverse problems
- Scalable and efficient algorithms for the propagation of uncertainty from data through inference to prediction for large-scale problems, with application to flow of the antarctic ice sheet
- Hierarchical Matrix Approximations of Hessians Arising in Inverse Problems Governed by PDEs
- Scalable Matrix-Free Adaptive Product-Convolution Approximation for Locally Translation-Invariant Operators
- Likelihood approximation with hierarchical matrices for large spatial datasets
- Randomized GPU Algorithms for the Construction of Hierarchical Matrices from Matrix-Vector Operations
- A fast randomized eigensolver with structured LDL factorization update
- Robust and accurate stopping criteria for adaptive randomized sampling in matrix-free hierarchically semiseparable construction
- PyAlbany: a Python interface to the C++ multiphysics solver Albany
Cited In (3)
This page was built for publication: Hierarchical off-diagonal low-rank approximation of Hessians in inverse problems, with application to ice sheet model initialization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6166000)
