Robust Physics-Guided Diffusion for Full-Waveform Inversion

2026-03-17

AI Summary

This work addresses the instability in full-waveform inversion caused by amplitude imbalance and phase misalignment by proposing a physics-guided diffusion inversion framework. The method integrates score-based generative priors with likelihood guidance derived from wave-equation modeling and introduces several key innovations: a Wasserstein-2 metric–based data consistency potential, wavefield enhancement through bounded weighting and observation-dependent normalization, and a preconditioned reverse diffusion mechanism that dynamically adjusts both guidance strength and spatial scale. Experiments on the OpenFWI dataset demonstrate that, under identical computational budgets, the proposed approach significantly outperforms conventional deterministic optimization and standard diffusion posterior sampling, yielding more stable and higher-quality reconstructions of subsurface structures.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning with Manifolds

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

We develop a robust physics-guided diffusion framework for full-waveform inversion that combines a score-based generative prior with likelihood guidance computed through wave-equation simulations. We adopt a transport-based data-consistency potential (Wasserstein-2), incorporating wavefield enhancement via bounded weighting and observation-dependent normalization, thereby improving robustness to amplitude imbalance and time/phase misalignment. On the inference side, we introduce a preconditioned guided reverse-diffusion scheme that adapts the guidance strength and spatial scaling throughout the reverse-time dynamics, yielding a more stable and effective data-consistency guidance step than standard diffusion posterior sampling (DPS). Numerical experiments on OpenFWI datasets demonstrate improved reconstruction quality over deterministic optimization baselines and standard DPS under comparable computational budgets.

Problem

• Full-Waveform Inversion
• Amplitude Imbalance
• Time/Phase Misalignment
• Data Consistency
• Robust Reconstruction

Innovation

• physics-guided diffusion
• full-waveform inversion
• Wasserstein-2 potential
• guided reverse-diffusion
• score-based generative model