Program introduction
This program introduces machine learning techniques tailored to seismic data processing, with an emphasis on practical implementation, physical validity, and applications relevant to current industry problems.
Program scope
The training connects classical physical wavefield operators with modern data-driven methods, including deep learning and hybrid physics-informed neural networks (PINNs) that respect the underlying wave equations.
Program topics
Core thematic pillars covered throughout the training:
Denoising & Interpolation
Deep learning architectures for random/coherent noise attenuation and spatial data reconstruction.
Deep Learning Architectures
CNNs, Transformers, and autoencoders adapted for 2D/3D seismic waveform interpretation.
Physics-Informed Neural Networks
Wave equation constraints built into neural networks to keep results physically consistent.
Industry & Research Case Studies
Real deployment scenarios across offshore exploration, DAS monitoring, and microseismic tracking.
Courses & teaching materials
Resources are organized to build from foundational knowledge up to reproducible workflows on complex 2D/3D seismic datasets.
Foundational module
Introduction to Machine Learning for Seismology
Course notes (PDF), lecture slides, Python environment setup
Core hands-on
Deep Learning for Seismic Denoising & Interpolation
Course notes (PDF), PyTorch implementation exercises, benchmark datasets
Advanced workshop
Neural Network Architectures for Waveform Data
Course notes (PDF), Jupyter notebooks, pre-trained models
Research seminar
Physics-Informed & Hybrid Learning Approaches
Course notes (PDF), PINNs code examples, synthetic case studies
Capstone
Research Case Studies in ML-Based Processing
Technical documentation (PDF), field dataset benchmarks
Target audience
Graduate students, geophysicists, PhD researchers, and industry professionals looking to integrate machine learning into seismic acquisition, processing, and interpretation pipelines.
By clicking "Join", you agree to our Terms & Conditions