Starts: March 22, 18h00 EDT.
Advancing a framework to complement cognitive analysis / interpretation with ML methods to support regional to national scale geological mapping.
This engagement requires prior experience building ML projects in Python. Some understanding of landscapes and geology is helpful. Most data sets have already been prepared for a test area.
Weekly Meeting link: https://meet.jit.si/GeologicalmappingwithML
Brief statement of the problem
Leads / Advisors
Hazen Russell | Lead
Sedimentologist GS
William Parkinson | Lead
Technical Product Manager, EarthDaily Analytics
~Nicolas Benoit | Lead
Hydrogeologist, GSC
Overview
- Challenge - mapping geological materials that can have overlapping spectral signals and landscape positions. Extraction of irregular landform shapes and sizes that have inconsistent composition, moisture regime, and vegetation.
- Tools - Machine Learning in Python using scikit-learn, TensorFlow or PyTorch algorithms (classification, regression, clustering, Convolutional Neural Network and more) and Geostatistic (spatial covariance, interpolation, simulation, scaling and uncertainty characterization).
- The goal is to build an understanding of the challenge of replacing abstract cognitive analysis with ML analysis.
- Familiarity with environmental data sets such as digital elevation models (DEM), remote sensing data (LandSat), and geological landscapes and geological materials.
- Familiarity with computer programming (preferably Python), general knowledge with Machine Learning and Geostatistics.
- We hope to document and contribute to published workshop results.
- The specific minimal goal to be announced soon.
Tentative Timeline
# | Major Milestones | Expected time to finish |
1 | Get familiar with concepts and datasets. | 2 weeks |
2 | Data review and integration | 2-3 week |
3 | Get to a baseline model | 2 weeks |
4 | Hyper Parameter optimization | 2 weeks |
Our Journey - Geological Mapping with ML
Why join?
Aggregate Intellect hosts one of the most diverse ML communities in the world. Over the course of the working group:
- you’ll get an immersion into that community & walk out with some cool new friends;
- the first three participants / teams to reach the minimal goal will receive a $500 gift each. The remaining participants reaching the minimal goal will have ~$1000 divided amongst them. (A nice little push to encourage you to learn 😉);
- learn about environmental landscape analysis and ML.
The button below will take you to our coordination Slack channel. Just jump in, introduce yourself, and say you wanna join! We’ll try to help bring you up to speed.