Machine-Learning Seismic Processing Tasks by Fine Tuning a Pre-Trained Attention-Based Neural Network: Storseismic

Abstract
Multi-dimensional seismic data often contain signatures and geometrical features that can help processing of such data provide valuable subsurface information. We propose a framework to first learn (and store) those features in a pre-trained neural network model, we refer to as StorSeismic. We then use this network for specific seismic processing tasks in an efficient fine-tuning stage. We use the Bidirectional Encoder Representation from Transformers (BERT), utilized in natural language processing (NLP), for the pre-training and fine-tuning stages. The traces, as opposed to words in NLP, are randomly masked to allow the network to learn the structure of the shot gathers. We demonstrate this approach on synthetic data resembling a marine setting, and we will share real data applications in the workshop.

Citation
Alkhalifah, T., & Harsuko, R. (2022). Machine-Learning Seismic Processing Tasks by Fine Tuning a Pre-Trained Attention-Based Neural Network: Storseismic. 83rd EAGE Annual Conference & Exhibition Workshop Programme. https://doi.org/10.3997/2214-4609.202211018

Publisher
European Association of Geoscientists & Engineers

Conference/Event Name
83rd EAGE Annual Conference & Exhibition Workshop Programme

DOI
10.3997/2214-4609.202211018

Additional Links
https://www.earthdoc.org/content/papers/10.3997/2214-4609.202211018

Permanent link to this record