Physics-guided deep learning for seismic inversion: hybrid training and uncertainty analysis
Jian Sun, Kristopher A. Innanen, Chao Huang
The determination of subsurface elastic property models is crucial in quantitative seismic data processing and interpretation. This problem is commonly solved by deterministic physical methods, such as tomography or full waveform inversion (FWI). However, these methods are entirely local, and require accurate initial models. Deep learning represents a plausible class of methods for seismic inversion, which may avoid some of the issues of purely descent-based approaches. However, any generic deep learning network capable of relating each elastic property cell value to each sample in a seismic dataset would require a very large number of degrees of freedom. We propose a hybrid network design, involving both deterministic, physics-based modelling and data-driven deep learning components. From an optimization standpoint, both a data-driven model misfit (i.e., standard deep learning), and now a physics-guided data residual (i.e., a wave propagation network), are simultaneously minimized during the training of the network. An experiment is carried out to analyze the trade-off between two types of losses. Synthetic velocity building is employed to examine the capacity of hybrid training.