Accelerated Full Waveform Inversion by Deep Compressed Learning
Name
sensors-26-01832-v2.pdf
Size
1.47 MB
Format
Adobe PDF
Checksum (MD5)
9d6854442e94678009883a23710161c8
Author(s)
Gelboim, Maayan
Adler, Amir
Araya-Polo, Mauricio
Date Issued
March 13, 2026
Journal
Sensors
Publisher
Multidisciplinary Digital Publishing Institute
Citation
Gelboim, M., Adler, A., & Araya-Polo, M. (2026). Accelerated Full Waveform Inversion by Deep Compressed Learning. Sensors, 26(6), 1832.
Version
Final published version
Abstract
We propose and test a method to reduce the dimensionality of Full Waveform Inversion (FWI) inputs as a computational cost mitigation approach. Given modern seismic acquisition systems, the data (as an input for FWI) required for an industrial-strength case is in the teraflop level of storage; therefore, solving complex subsurface cases or exploring multiple scenarios with FWI becomes prohibitive. The proposed method utilizes a deep neural network with a binarized sensing layer that learns by compressed learning seismic acquisition layouts from a large corpus of subsurface models. Thus, given a large seismic data set to invert, the trained network selects a smaller subset of the data, then by using representation learning, an autoencoder computes latent representations of the shot gathers, followed by K-means clustering of the latent representations to further select the most relevant shot gathers for FWI. This approach can effectively be seen as a hierarchical selection. The proposed approach consistently outperforms random data sampling, even when utilizing only 10% of the data for 2D FWI, and these results pave the way to accelerating FWI in large scale 3D inversion.
MIT Department
Massachusetts Institute of Technology. Department of Mathematics
Center for Brains, Minds, and Machines
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.3390/s26061832
