scholarly journals Probabilistic inverse problems using machine learning - applied to inversion of airborne EM data.

2021 ◽  
Author(s):  
Thomas Hansen
Keyword(s):  
Machine Learning ◽  
Inverse Problems ◽  
Airborne Em

Design Methods and Strategies for Forward and Inverse Problems of Turbine Blades Based on Machine Learning

2022 ◽  
Vol 31 (1) ◽  
pp. 82-95
Author(s):  
Haimeng Zhou ◽  
Kaituo Yu ◽  
Qiao Luo ◽  
Lei Luo ◽  
Wei Du ◽  
...  
Keyword(s):  
Machine Learning ◽  
Inverse Problems ◽  
Turbine Blades ◽  
Design Methods

Inverse Problems of Heterogeneous Geological Layers Exploration Seismology Solution by Methods of Machine Learning

2021 ◽  
Vol 42 (7) ◽  
pp. 1728-1737
Author(s):  
M. V. Muratov ◽  
V. V. Ryazanov ◽  
V. A. Biryukov ◽  
D. I. Petrov ◽  
I. B. Petrov
Keyword(s):  
Machine Learning ◽  
Inverse Problems ◽  
Exploration Seismology

Forward-inverse modeling based on scalar and vector radiative transfer models for coupled atmosphere-surface systems and machine learning tools

2020 ◽  
Author(s):  
Knut Stamnes ◽  
Børge Hamre ◽  
Snorre Stamnes ◽  
Nan Chen ◽  
Yongzhen Fan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Inverse Problem ◽  
Radiative Transfer ◽  
Inverse Problems ◽  
Surface Properties ◽  
Inverse Modeling ◽  
Atmospheric Correction ◽  
Machine Learning Techniques ◽  
Ocean Bottom ◽  
Curvature Effects

<p>Reliable retrieval of atmospheric and surface properties from sensors deployed on satellite platforms rely on accurate simulations of the electromagnetic (EM) signal measured by such sensors. A forward radiative transfer (RT) model of the coupled atmosphere-surface system can be used to simulate how the EM signal responds to changes in atmospheric and surface properties. Realistic RT modeling is a prerequisite for solving the inverse problem, i.e. to infer atmospheric and surface parameters from the EM signals measured at the top of the atmosphere. The surface may consist of a soil-plant canopy, a snow/ice covered surface or an open water body (ocean, lake, river system). An overview will be provided of forward and inverse RT in such coupled atmosphere-surface systems. A coupled system consisting of two adjacent slabs separated by an interface across which the refractive index changes abruptly from its value in air to that in water /ice [1] will be used as an example. Several examples of how to formulate and solve inverse problems involving coupled atmosphere-water systems [2] will be provided to illustrate how solutions to the RT equation can be used as a forward model to solve practical inverse problems. Cloud screening [3], atmospheric correction [4], treatment of two-dimensional surface roughness, Earth curvature effects, and ocean bottom reflection for shallow water in coastal areas will be discussed, and the advantage of using powerful machine learning techniques to solve the inverse problem will be emphasized.</p><p><strong>References</strong></p><p>[1] Stamnes, K., and J. J. Stamnes, <em>Radiative Transfer in Coupled Environmental Systems</em>, , 2015.</p><p>[2] Stamnes, K., B. Hamre, S. Stamnes, N. Chen, Y. Fan, W. Li, Z. Lin, and J. J. Stamnes, Progress in forward-inverse modeling based on radiative transfer tools for coupled atmosphere-snow/ice-ocean systems: A review and description of the AccuRT model, , 8, 2682, 2018.</p><p>[3] Chen N., W. Li, C. Gatebe, T. Tanikawa, M. Hori, R. Shimada; T. Aoki, and K. Stamnes, New cloud mask algorithm based on machine learning methods and radiative transfer simulations, , 219, 62-71, 2018.</p><p>[4] Fan, Y., W. Li, C. K. Gatebe, C. Jamet, G. Zibordi, T. Schroeder, and K. Stamnes, Atmospheric correction and aerosol retrieval over coastal waters using multilayer neural networks, , 199, 218-240, 2017.</p>

scholarly journals Efficient probabilistic inversion using the rejection sampler—exemplified on airborne EM data

2020 ◽  
Vol 224 (1) ◽  
pp. 543-557
Author(s):  
Thomas M Hansen
Keyword(s):  
Inverse Problems ◽  
Posterior Distribution ◽  
Prior Information ◽  
Local Minima ◽  
Electromagnetic Data ◽  
Novel Approach ◽  
Airborne Electromagnetic ◽  
Airborne Electromagnetic Data ◽  
Airborne Em ◽  
Probabilistic Inversion

SUMMARY Probabilistic inversion methods, typically based on Markov chain Monte Carlo, exist that allow exploring the full uncertainty of geophysical inverse problems. The use of such methods is though limited by significant computational demands, and non-trivial analysis of the obtained set of dependent models. Here, a novel approach, for sampling the posterior distribution is suggested based on using pre-calculated lookup tables with the extended rejection sampler. The method is (1) fast, (2) generates independent realizations of the posterior, and (3) does not get stuck in local minima. It can be applied to any inverse problem (and sample an approximate posterior distribution) but is most promising applied to problems with informed prior information and/or localized inverse problems. The method is tested on the inversion of airborne electromagnetic data and shows an increase in the computational efficiency of many orders of magnitude as compared to using the extended Metropolis algorithm.

Accelerating inverse problems in seismology using adjoint-based machine learning

2021 ◽  
Author(s):  
Lars Gebraad ◽  
Sölvi Thrastarson ◽  
Andrea Zunino ◽  
Andreas Fichtner
Keyword(s):  
Machine Learning ◽  
Monte Carlo ◽  
Bayesian Inference ◽  
Inverse Problems ◽  
Uncertainty Quantification ◽  
Neural Nets ◽  
Monte Carlo Algorithms ◽  
Gradient Based ◽  
Speed Up ◽  
Machine Learning Models

<p><span>Uncertainty quantification is an essential part of many studies in Earth science. It allows us, for example, to assess the quality of tomographic reconstructions, quantify hypotheses and make physics-based risk assessments. In recent years there has been a surge in applications of uncertainty quantification in seismological inverse problems. This is mainly due to increasing computational power and the ‘discovery’ of optimal use cases for many algorithms (e.g., gradient-based Markov Chain Monte Carlo (MCMC). Performing Bayesian inference using these methods allows seismologists to perform advanced uncertainty quantification. However, oftentimes, Bayesian inference is still prohibitively expensive due to large parameter spaces and computationally expensive physics.</span></p><p><span>Simultaneously, machine learning has found its way into parameter estimation in geosciences. Recent works show that machine learning both allows one to accelerate repetitive inferences [e.g. </span>Shahraeeni & Curtis 2011, <span>Cao et al. 2020] as well as speed up single-instance Monte Carlo algorithms </span><span>using surrogate networks </span><span>[Aleardi 2020]. These advances allow seismologists to use machine learning as a tool to bring accurate inference on the subsurface to scale.</span></p><p>In this work, we propose the novel inclusion of adjoint modelling in machine learning accelerated inverse problems. The aforementioned references train machine learning models on observations of the misfit function. This is done with the aim of creating surrogate but accelerated models for the misfit computations, which in turn allows one to compute this function and its gradients much faster. This approach ignores that many physical models have an adjoint state, allowing one to compute gradients using only one additional simulation.</p><p>The inclusion of this information within gradient-based sampling creates performance gains in both training the surrogate and the sampling of the true posterior. We show how machine learning models that approximate misfits and gradients specifically trained using adjoint methods accelerate various types of inversions and bring Bayesian inference to scale. Practically, the proposed method simply allows us to utilize information from previous MCMC samples in the algorithm proposal step.</p><p>The application of the proposed machinery is in settings where models are extensively and repetitively run. Markov chain Monte Carlo algorithms, which may require millions of evaluations of the forward modelling equations, can be accelerated by off-loading these simulations to neural nets. This approach is also promising for tomographic monitoring, where experiments are repeatedly performed. Lastly, the efficiently trained neural nets can be used to learn a likelihood for a given dataset, to which subsequently different priors can be efficiently applied.<span> We show examples of all these use cases.</span></p><p> </p><p>Lars Gebraad, Christian Boehm and Andreas Fichtner, 2020: Bayesian Elastic Full‐Waveform Inversion Using Hamiltonian Monte Carlo.</p><p>Ruikun Cao, Stephanie Earp, Sjoerd A. L. de Ridder, Andrew Curtis, and Erica Galetti, 2020: Near-real-time near-surface 3D seismic velocity and uncertainty models by wavefield gradiometry and neural network inversion of ambient seismic noise.</p><p>Mohammad S. Shahraeeni and Andrew Curtis, 2011: Fast probabilistic nonlinear petrophysical inversion.</p><p><span>Mattia Aleardi, 2020: Combining discrete cosine transform and convolutional neural networks to speed up the Hamiltonian Monte Carlo inversion of pre‐stack seismic data.</span></p>

scholarly journals Regularization, Bayesian Inference, and Machine Learning Methods for Inverse Problems

Entropy ◽  
2021 ◽  
Vol 23 (12) ◽  
pp. 1673
Author(s):  
Ali Mohammad-Djafari
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Inverse Problems ◽  
Data Model ◽  
Hidden Variables ◽  
Model Matching ◽  
Regularization Theory ◽  
Probability Models ◽  
Map Interpretation ◽  
And Inversion

Classical methods for inverse problems are mainly based on regularization theory, in particular those, that are based on optimization of a criterion with two parts: a data-model matching and a regularization term. Different choices for these two terms and a great number of optimization algorithms have been proposed. When these two terms are distance or divergence measures, they can have a Bayesian Maximum A Posteriori (MAP) interpretation where these two terms correspond to the likelihood and prior-probability models, respectively. The Bayesian approach gives more flexibility in choosing these terms and, in particular, the prior term via hierarchical models and hidden variables. However, the Bayesian computations can become very heavy computationally. The machine learning (ML) methods such as classification, clustering, segmentation, and regression, based on neural networks (NN) and particularly convolutional NN, deep NN, physics-informed neural networks, etc. can become helpful to obtain approximate practical solutions to inverse problems. In this tutorial article, particular examples of image denoising, image restoration, and computed-tomography (CT) image reconstruction will illustrate this cooperation between ML and inversion.

Sign in / Sign up

Export Citation Format

Share Document