Magnetotelluric inversion using supervised learning trained with random smooth geoelectric models

Liu，Lian1; Yang，Bo2; Xu，Yixian2; Yang，Dikun1

10.1190/image2023-3910431.1

2023-12-14

1052-3812

1949-4645

SEG Technical Program Expanded Abstracts

2023-August

489-492

Supervised learning has emerged as an effective approach to solving geophysical inversions. The quality of such inversion depends mainly on the training datasets (labeled samples), usually created by synthetic models and forward simulations. We propose a straightforward approach to generating abundant representative labeled samples for 2D or 3D inversion of magnetotelluric (MT) data (or other similar exploration methods). The approach requires very little prior information and has been demonstrated to be effective for the quick recovery of random and blocky models.

第一 ; 通讯

英语

EI

20235215283737

Supervised learning

Geophysics:481.3

2-s2.0-85180531299

Scopus

1.Department of Earth and Space Sciences,Southern University of Science and Technology,China
2.School of Earth Sciences,Zhejiang University,China

Liu，Lian,Yang，Bo,Xu，Yixian,et al. Magnetotelluric inversion using supervised learning trained with random smooth geoelectric models[C],2023:489-492.