Machine Learning Methods in GEOScience Short Course: April 4, 2025

The short course Machine Learning Methods in Geoscience consists of lectures that explain the high-level concepts underlying fully connected Neural Networks, Convolutional Neural networks (CNN), clustering methods, Recurrent Neural Networks, and ChatGPT. It is intended for practicing scientists and engineers in geosciences and engineering who have little knowledge of ML methods but desire a hands-on introduction to Machine Learning (ML) methods and their applications to scientific problems. The goal is to learn the basics of some ML methods and their implementation on free software that can be used for solving real-world problems.

Computer exercises consist of cluster identification of geochemical signatures for rock formations associated with hominin fossils near Lake Turkana, CNN identification of bird types from photographs, CNN detection of cracks along a fractured dam face from drone photos, and discrimination of destructive weeds from plants in drone photos near Kennecott mine.

The course will be delivered April 4th at the EGI offices (8:30-4:30 PM MST). For in-person attendees, a laptop with MATLAB installed is preferred or a limited number of workstations with MATLAB will be available at EGI. The topics are listed below. Sessions will be based on Schuster’s recent book: Machine Learning Methods in Geoscience (library.seg.org/doi/10.1190/1.9781560804048).

This book is included with course registration. A digital version will be provided to registrants taking the course online.

Unsupervised Machine Learning

Overview of Course and Machine Learning Methods: 8:30-9:20

Unsupervised Clustering: K-Means Clustering: 9:20-9:40

Colab Lab: K-Means Clustering of Yellowstone Old Faithful Earthquake Events: 9:40-10:00

Unsupervised Clustering: DBSCAN & PCA: 10:10-10:40

Colab Lab: DBSCAN China Earthquakes: 10:40-11:00

Supervised Clustering: Gaussian Discrimination Analysis: 11:50-12:15

MATLAB: Lake Turkana-Africa Geochemical Data and Rock Formation Identification: 12:15-12:30

MATLAB: Distinguishing Weeds from Plants from Multispectral Photos: 12:15-12:30 (Optional)

Cost: $750 USD | Book (physical or e-edition) is included.

30 in-person and 30 online spots available

REGISTER AND PAY | USE PROMO CODE 50OFF