EOSC 510 (Data Analysis in Atmospheric, Earth and Ocean Sciences)

This is an online graduate course on applying machine learning and statistical methods to environmental sciences.

Course description and outline.    Textbook (optional)

Week 1 (Sept. 3-8):

Chapter 0 (Introduction and course setup)

Chapter 1. Correlation and regression

Week 2 (Sept. 9-15):

*Homework #1 assigned (due Sept.24).

Chapter 2. Principal component analysis (PCA) and rotated PCA

Week 3 (Sept. 16-22):

Week 4 (Sept.23-29):

Chapter 3. Canonical correlation analysis (CCA)

*Homework #2 assigned (due Oct.10).

Chapter 4. Time series

Week 5 (Sept.30-Oct.6):

Week 6 (Oct.7-Oct.13):

Chapter 5. Classification and clustering

Week 7 (Oct.14-Oct.20):

*Homework #3 assigned (due Oct.29).

Chapter 6. Feed-forward neural network models

Week 8 (Oct.21-Oct.27):

Midterm exam: Tuesday, 22 Oct. 11:00-12:15 (Pacific time)

The exam covers course material up to the end of Week 6 (i.e. Classification and clustering).

Week 9 (Oct.28-Nov.3):

Chapter 7. Nonlinear optimization

Week 10 (Nov.4-10):

Chapter 8. Learning and generalization

Week 11 (Nov.11-17):

*Homework #4 assigned (due Dec.9).

Chapter 9. Tree-based methods

Week 12 (Nov.18-24):

Chapter 10. Forecast verification

Week 13 (Nov.25-Dec.1):

Chapter 11. Nonlinear principal component analysis (NLPCA)

Chapter 12. Kernel methods (optional material)


Final exam: Friday, 6 Dec. 12:00-14:30 (Pacific time)

The exam covers the course material taught over the whole term, but is weighted more heavily towards material taught after the midterm exam.