---
title: "Continually Learning Neural-Operator Surrogate for 3D Airborne Electromagnetic Bayesian Inversion"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.25932v1"
permalink: "https://feed.astralor.com/items/itm_a8cd29aed4a00538ed9583df"
published: "2026-08-26T15:43:10.000Z"
score: 70
channel: "ai"
tags: ["Continual Learning","Mineral Exploration","Neural Operator","Airborne Electromagnetic (AEM)","Bayesian Inversion"]
---

# Continually Learning Neural-Operator Surrogate for 3D Airborne Electromagnetic Bayesian Inversion

## AI commentary

> The real highlight is grounding continual learning in a concrete geophysical inversion scenario, allowing the surrogate to improve across projects. For computational geophysics and mineral exploration teams, this is a promising acceleration path; for AI-for-science researchers, it demonstrates how ensemble disagreement checks and solver fallback can ensure reliability.

## At a glance

The paper tackles the prohibitive forward-solve cost in three-dimensional probabilistic inversion of time-domain airborne electromagnetic (AEM) data. The authors develop a continually learning neural-operator surrogate that replaces the solver inside Bayesian inversion. They leverage the invariance of Maxwell's laws to avoid dependence on a single geological prior, and use continual learning across consecutive priors so the surrogate improves over future case studies. A validity check built on ensemble disagreement diverts out-of-training-range measurements to the full solver. With this surrogate, the same Markov chain Monte Carlo sampler reproduces the full-solver posterior, and its credible intervals cover the truth within 2.6 percentage points. Applied to the 2013 Capricorn TEMPEST survey in Western Australia, the surrogate inverts over two million soundings in seconds, a computation previously infeasible for the solver. The framework delivers uncertainty-quantified conductivity imaging at survey scale, which is considered essential for near real-time mineral-systems targeting.

## Key points

- Proposes a continually learning neural-operator surrogate that replaces the 3D AEM forward solver and accumulates knowledge across priors.
- Uses ensemble disagreement as a validity check to route out-of-range measurements to the solver.
- Inverts over two million soundings from the 2013 Capricorn TEMPEST survey in seconds.
- The surrogate-driven MCMC posterior matches the full-solver posterior, with credible intervals covering the truth within 2.6 percentage points.

## Full text

Three-dimensional probabilistic inversion of time-domain airborne electromagnetic (AEM) data is limited by the cost of the forward solve. Even though one simulation takes only tens of seconds, a Bayesian inversion of a survey of millions of soundings requires of order $10^{10}$ forward evaluations. To address this, we develop a continually learning neural-operator surrogate of the three-dimensional AEM forward operator that replaces the solver inside the Bayesian inversion. We start from the point of view that regardless of what geological prior is specified, Maxwell's laws remain invariant. Secondly, we avoid the limitation of learning on a single prior by continual learning on consecutive priors, which means our surrogate becomes richer as it is applied in future case studies, either by the authors, or by the scientific community. We use a validity check built on ensemble disagreement to divert cases with measurements outside the training range to the solver. Driven by the surrogate, the identical Markov chain Monte Carlo sampler reproduces the full-solver posterior, and its credible intervals cover the truth within 2.6 percentage points. Applied to the 2013 Capricorn TEMPEST survey in Western Australia, the surrogate inverts over two million soundings in seconds, a computation infeasible for the solver. Testing the geological prior against the entire survey costs minutes. The framework delivers uncertainty-quantified conductivity imaging at survey scale, which we believe is essential to perform near real-time mineral-systems targeting with geophysics.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_a8cd29aed4a00538ed9583df