ML-descent: An optimization algorithm for full-waveform inversion using machine learning

TL;DR: A comprehensive review of the deep learning techniques being applied to seismic datasets, covering approaches, limitations, and opportunities is provided by Mousavi and Beroza as mentioned in this paper , who present a systematic overview of trends, challenges and opportunities in applications of deep learning methods in seismology.

TL;DR: An iterative deep neural network to solve full-wave inverse scattering problems in the 2D case and can achieve super-resolution reconstruction, meanwhile keeping the simulated data of reconstructed models well consistent with the observed data.

TL;DR: Applications of deep neural networks in exploration seismology involve seismic data acquisition, preprocessing, processing, interpretations, and modeling tasks. DNNs have revolutionized exploration seismology but face challenges with out-of-distribution generalization and interpretability.

TL;DR: Physics-informed neural networks for seismic wave modeling in semi-infinite domain can accurately predict wave propagation without labeled data. The model incorporates absorbing boundary conditions and sequential training strategies to handle large domain sizes and parametric loading.

TL;DR: In this article, an effective way of initializing the weights that allows deep autoencoder networks to learn low-dimensional codes that work much better than principal components analysis as a tool to reduce the dimensionality of data is described.

TL;DR: The numerical tests indicate that the L-BFGS method is faster than the method of Buckley and LeNir, and is better able to use additional storage to accelerate convergence, and the convergence properties are studied to prove global convergence on uniformly convex problems.

TL;DR: This chapter discusses Monte Carol methods, the least-absolute values criterion and the minimax criterion, and their applications to functional inverse problems.