Structure-aware Bayesian inversion improves subsurface imaging with diffusion models
Structure-aware Bayesian full waveform inversion via diffusion-regularized Stein variational gradient descent
Machine Learning
Summary
Finding underground structures accurately is hard when starting points are off or data is messy. The authors use a technique that keeps many possible models at once to understand uncertainty better. They add a smart tool that learns realistic underground shapes to guide these models, making the results more believable and less noisy. Their method helps create clearer pictures of underground structures and better shows where uncertainty is highest.
What this means in practice
- •For oil and gas exploration teams: Produce more accurate underground velocity models with clearer geological structures for improved resource exploration using Bayesian inversion guided by learned priors.
- •For geotechnical engineering firms: Generate reliable subsurface models with uncertainty estimates in complex geological areas for safer construction planning and risk assessment.
Authors
Chao Li, Mrinal K. Sen, Sergey Fomel, Yangkang Chen
Abstract
Full waveform inversion (FWI) is a powerful tool for high-resolution subsurface model reconstruction, but it remains highly nonlinear and ill-posed, especially when the initial model is inaccurate or the acquired data are noisy and incomplete. Annealed Stein variational gradient descent (ASVGD) provides a particle-based Bayesian framework for FWI by approximating the posterior distribution with an ensemble of velocity models, thereby enabling uncertainty quantification beyond a single deterministic solution. However, conventional ASVGD-FWI commonly relies on simple handcrafted priors, such as Gaussian or smoothing assumptions, which are often insufficient to represent complex geological structures. To address this limitation, we introduce a learned diffusion model (DM) as an implicit regularization mechanism within the ASVGD-FWI framework. Instead of explicitly evaluating the diffusion prior density, the trained diffusion model acts as a learned geological projection operator that guides updated particles toward the distribution of plausible velocity models. In this way, the proposed method combines data consistency from FWI, posterior diversity from ASVGD, and structural realism from the diffusion prior. Numerical experiments demonstrate that the diffusion-regularized ASVGD-FWI produces more geologically consistent mean models, suppresses nonphysical artifacts, and yields uncertainty estimates that are more concentrated around structurally complex and poorly constrained regions. These results suggest that learned diffusion priors can effectively enhance Bayesian FWI by improving both inversion stability and the interpretability of posterior uncertainty.