WISER: multimodal variational inference for full-waveform inversion without dimensionality reduction

TL;DR

WISER introduces a semi-amortized variational inference method for 2D full-waveform inversion, enabling efficient and reliable uncertainty quantification of subsurface models without dimensionality reduction.

Contribution

The paper presents WISER, a novel framework combining amortized and non-amortized inference for accurate uncertainty estimation in full-waveform inversion.

Findings

WISER provides full-resolution uncertainty estimates.

The method is computationally feasible for 2D models.

WISER effectively captures multimodal posterior distributions.

Abstract

We present a semi-amortized variational inference framework designed for computationally feasible uncertainty quantification in 2D full-waveform inversion to explore the multimodal posterior distribution without dimensionality reduction. The framework is called WISER, short for full-Waveform variational Inference via Subsurface Extensions with Refinements. WISER leverages the power of generative artificial intelligence to perform approximate amortized inference that is low-cost albeit showing an amortization gap. This gap is closed through non-amortized refinements that make frugal use of acoustic wave physics. Case studies illustrate that WISER is capable of full-resolution, computationally feasible, and reliable uncertainty estimates of velocity models and imaged reflectivities.