Instructor: Michael Pyrcz
PGE 379 / 383 – Subsurface Machine Learning Unique I.D.: PGE 379 18850 / PGE 383 TBD
“You will learn the theory and practice of data analytics and machine learning for subsurface resource modeling”.
We will build from the ground up covering fundamental probability and statistics; inference (clustering, multidimensional scaling) and prediction (regression, decision trees, random forest, support vector machines and neural nets); model training, testing, tuning and validation; and workflows for integrating many features; workflows for anomaly detection, spatiotemporal prediction, and uncertainty integration and communicating results within a project team to impact subsurface decision making.
The course is cross listed as an undergraduate (PGE 379) and graduate course (PGE 383). Lectures are the same, the assignment will vary for undergraduate and graduate variants.
Course Online: The course lectures are recorded on posted on YouTube and the example workflows are available on GitHub for anyone to follow along.
Prerequisites: Admission to an appropriate major sequence in engineering, geoscience or consent of instructor. Students must be willing to learn and use Basic Python and Statistical and Machine Learning Python Packages to complete assignments and participate in class. Basic Python use will be covered in the first couple of lectures and well-documented workflows and hands-on will be provided to support the students (Jupyter Notebooks).
MEETS: MWF 11:00am – 12:00am, at CPE 2.202
INSTRUCTOR: Michael J. Pyrcz e-mail YouTube GitHub
TEACHING ASSISTANT: TBD
READINGS: There is no course textbook. The provided notes and example workflows are comprehensive, but students interested in additional reading are welcome to refer to:
- Machine Learning:
Hastie, T, Tibshirani, R., and Friedman, J., 2012, The Elements of Statistical Learning; Data Mining, Inference and Prediction, Springer.
- Subsurface Data Analytics and Modeling:
Pyrcz, M. and Deutsch, C., Geostatistical Reservoir Modeling, Oxford University Press, New York, 2014.
- Also, various journal papers will be posted for reference.
GRADING: Assignments 25% Quizzes 20% Midterms 25% Final 25% Class participation, Quizzes 5%
Note: the final will be an optional, no-risk final. Students may choose to write it to improve their course grade. If the result of the final is lower than the pre-final term result, it will be thrown out.
COURSE OBJECTIVES:
You will gain:
- Data Analytics: fundamental probability, statistics, data preparation, summarization and visualization prerequisites for modeling
- Spatiotemporal Inference: approaches for learning from data, discovering and communicating relationships
- Spatiotemporal Prediction: knowledge and experience with a variety of parametric and nonparametric models
- Model Checking: methods to train, test and tune subsurface models
- Uncertainty Modeling: methods to quantify, integrate and summarize uncertainty
- Communicating Model Results: maximizing the impact of your modeling work by communicating modeling results to inform decision making
- Practical Workflows: a wide variety of practical workflows that you will build from to kick-start your successful career in the subsurface digital revolution.
This class should be useful to those interested to learn about machine learning and how to apply it to the subsurface.
I have other resources available online, check out my other Python demonstrations and Python Geostatistics package GeostatsPy.
I hope that this is helpful to those that want to learn more about geostatistics, subsurface modeling, data analytics and machine learning.
Want to invite me to visit your company for training, mentoring, project review, workflow design and consulting, I'd be happy to drop by and work with you!
Interested in partnering, supporting my graduate student research or my Subsurface Data Analytics and Machine Learning consortium (co-PIs including Profs. Foster, Torres-Verdin and van Oort)? My research combines data analytics, stochastic modeling and machine learning theory with practice to develop novel methods and workflows to add value. We are solving challenging subsurface problems!
I can be reached at mpyrcz@austin.utexas.edu.
I'm always happy to discuss,
Michael
Michael Pyrcz, Ph.D., P.Eng. Associate Professor The Hildebrand Department of Petroleum and Geosystems Engineering, Bureau of Economic Geology, The Jackson School of Geosciences, The University of Texas at Austin