• Corpus ID: 203626777

Neural network augmented wave-equation simulation

  title={Neural network augmented wave-equation simulation},
  author={Ali Siahkoohi and Mathias Louboutin and F. Herrmann},
Accurate forward modeling is important for solving inverse problems. An inaccurate wave-equation simulation, as a forward operator, will offset the results obtained via inversion. In this work, we consider the case where we deal with incomplete physics. One proxy of incomplete physics is an inaccurate discretization of Laplacian in simulation of wave equation via finite-difference method. We exploit intrinsic one-to-one similarities between timestepping algorithm with Convolutional Neural… 
Numerical wave propagation aided by deep learning
A deep learning approach for wave propagation in media with multiscale wave speed, using a second-order linear wave equation model, and it is demonstrated that the trained neural networks can approximate the nonlinear dependence of the propagation on the wave speed as long as the causality is appropriately sampled in training data.
Solving the acoustic VTI wave equation using physics-informed neural networks
Frequency-domain wavefield solutions corresponding to the anisotropic acoustic wave equations can be used to describe the anisotropic nature of the earth. To solve a frequency-domain wave equation,
Solving the wave equation with physics-informed deep learning
We investigate the use of Physics-Informed Neural Networks (PINNs) for solving the wave equation. Whilst PINNs have been successfully applied across many physical systems, the wave equation presents
Solving the frequency-domain acoustic VTI wave equation using physics-informed neural networks
Frequency-domain wavefield solutions corresponding to the anisotropic acoustic wave equation can be used to describe the anisotropic nature of the Earth. To solve a frequency-domain wave equation,
Deep-neural-networks-based approaches for Biot–squirt model in rock physics
A new cost-effective surrogate model using deep neural network (DNN) for seismic wave propagation in rocks saturated with fluid is presented and reveals that the fast wave-field simulation can be implemented once the results with lower accuracy are obtained.
Trend to equilibrium for the kinetic Fokker-Planck equation via the neural network approach
Deep learning in photoacoustic tomography: current approaches and future directions
A review of the literature on learned image reconstruction, summarizing the current trends and explaining how these approaches fit within, and to some extent have arisen from, a framework that encompasses classical reconstruction methods shows how these techniques can be understood from a Bayesian perspective, providing useful insights.
j-Wave: An open-source differentiable wave simulator
We present an open-source differentiable acoustic simulator, j-Wave, which can solve time-varying and time-harmonic acoustic problems. It supports automatic differentiation, which is a program
Accelerating innovation with software abstractions for scalable computational geophysics
The SLIM open-source software framework for computational geophysics, and more generally, inverse problems based on the wave-equation, is presented, aimed at scalable research and development by designing multiple layers of abstractions.
Frequency-domain wavefield solutions corresponding to the anisotropic acoustic wave equations can be used to describe the anisotropic nature of the earth. To solve a frequency-domain wave equation, we


Fast approximate simulation of seismic waves with deep learning
This work simulates the response of acoustic seismic waves in horizontally layered media using a deep neural network that is able to directly approximate the recorded seismic response at multiple receiver locations in a single inference step, resulting in an order of magnitude reduction in simulation time.
Devito: an embedded domain-specific language for finite differences and geophysical exploration
Devito, a new domain-specific language for implementing high-performance finite-difference partial differential equation solvers within Python and making heavy use of SymPy, a symbolic mathematics library, is introduced, making it possible to develop finite-Difference simulators quickly using a syntax that strongly resembles the mathematics.
Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
  • M. Raissi
  • Computer Science
    J. Mach. Learn. Res.
  • 2018
This work puts forth a deep learning approach for discovering nonlinear partial differential equations from scattered and potentially noisy observations in space and time by approximate the unknown solution as well as the nonlinear dynamics by two deep neural networks.
Learned Iterative Solvers for the Helmholtz Equation
This work proposes a ‘learned’ iterative solver for the Helmholtz equation, by combining traditional Krylov-based solvers with machine learning, and demonstrates the effectiveness of this approach under a 1.5-D assumption.
Adam: A Method for Stochastic Optimization
This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Architecture and Performance of Devito, a System for Automated Stencil Computation
The architecture of the Devito compiler, as well as the performance optimizations that are applied when generating code, are presented and the effectiveness of such performance optimizations is demonstrated using operators drawn from seismic imaging applications.
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
This work considers image transformation problems, and proposes the use of perceptual loss functions for training feed-forward networks for image transformation tasks, and shows results on image style transfer, where aFeed-forward network is trained to solve the optimization problem proposed by Gatys et al. in real-time.
How transferable are features in deep neural networks?
This paper quantifies the generality versus specificity of neurons in each layer of a deep convolutional neural network and reports a few surprising results, including that initializing a network with transferred features from almost any number of layers can produce a boost to generalization that lingers even after fine-tuning to the target dataset.
Generative Adversarial Nets
We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a
Deep Residual Learning for Image Recognition
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.