Seismic Full-Waveform Inversion Using Deep Learning Tools and Techniques

TL;DR

This paper presents a novel approach to seismic full-waveform inversion by reformulating it as a recurrent neural network and utilizing deep learning optimization techniques for faster convergence.

Contribution

It introduces a deep learning framework for seismic inversion, demonstrating improved efficiency and linking traditional adjoint methods with automatic differentiation.

Findings

Recurrent neural network formulation of seismic inversion.

Adam optimizer achieves quicker convergence than traditional methods.

Gradient calculation matches the adjoint state method.

Abstract

I demonstrate that the conventional seismic full-waveform inversion algorithm can be constructed as a recurrent neural network and so implemented using deep learning software such as TensorFlow. Applying another deep learning concept, the Adam optimizer with minibatches of data, produces quicker convergence toward the true wave speed model on a 2D dataset than Stochastic Gradient Descent and than the L-BFGS-B optimizer with the cost function and gradient computed using the entire training dataset. I also show that the cost function gradient calculation using reverse-mode automatic differentiation is the same as that used in the adjoint state method.