A stochastic gradient descent approach with partitioned-truncated singular value decomposition for large-scale inverse problems of magnetic modulus data
From MaRDI portal
Publication:5081796
Abstract: We propose a stochastic gradient descent approach with partitioned-truncated singular value decomposition for large-scale inverse problems of magnetic modulus data. Motivated by a uniqueness theorem in gravity inverse problem and realizing the similarity between gravity and magnetic inverse problems, we propose to solve the level-set function modeling the volume susceptibility distribution from the nonlinear magnetic modulus data. To deal with large-scale data, we employ a mini-batch stochastic gradient descent approach with random reshuffling when solving the optimization problem of the inverse problem. We propose a stepsize rule for the stochastic gradient descent according to the Courant-Friedrichs-Lewy condition of the evolution equation. In addition, we develop a partitioned-truncated singular value decomposition algorithm for the linear part of the inverse problem in the context of stochastic gradient descent. Numerical examples illustrate the efficacy of the proposed method, which turns out to have the capability of efficiently processing large-scale measurement data for the magnetic inverse problem. A possible generalization to the inverse problem of deep neural network is discussed at the end.
Recommendations
- Regularizing inversion of susceptibility with projection onto convex set using full tensor magnetic gradient data
- A spectral conjugate gradient method for nonlinear inverse problems
- An efficient iterative approach for large-scale separable nonlinear inverse problems
- scientific article; zbMATH DE number 4094696
- Regularized Linear Inversion with Randomized Singular Value Decomposition
- An analysis of stochastic variance reduced gradient for linear inverse problems *
- A level-set algorithm for the inverse problem of full magnetic gradient tensor data
- The alternating descent conditional gradient method for sparse inverse problems
- Modified truncated randomized singular value decomposition (MTRSVD) algorithms for large scale discrete ill-posed problems with general-form regularization
Cites work
- A level-set algorithm for the inverse problem of full magnetic gradient tensor data
- A Stochastic Approximation Method
- A variational level set approach to multiphase motion
- An improved fast local level set method for three-dimensional inverse gravimetry
- scientific article; zbMATH DE number 194472 (Why is no real title available?)
- Optimization for simulation: theory vs. practice
- Optimization methods for large-scale machine learning
- Robust Stochastic Approximation Approach to Stochastic Programming
- Simultaneously recovering both domain and varying density in inverse gravimetry by efficient level-set methods
Cited in
(2)
This page was built for publication: A stochastic gradient descent approach with partitioned-truncated singular value decomposition for large-scale inverse problems of magnetic modulus data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5081796)
