LEO6006: Data Science and Machine Learning for Aerospace Systems

Course Info.

Course descriptions

• This course provides a theoretical foundation for data science and machine learning. It covers key concepts, including regression, classification, probabilistic supervised learning, unsupervised learning, and optimization. Emphasis is placed on understanding the mathematical principles and theoretical frameworks underlying these techniques, with applications to aerospace domains.

Instructors

• Jong-Han Kim (Rm.507)

Lectures

• Tue/Thr 16:00-17:15 (Rm.217 @aerospace campus)

Office hours

• JHK: Tue/Thr 17:30-19:00 (Rm.507), or by appointments.

Prerequisites

• Previous exposure to linear algebra, probability, and programming.

• Working knowledge on optimization will be a plus.

Reference textbooks

• There are no required textbooks.

• Introduction to applied linear algebra - vectors, matrices, and least squares by Boyd and Vandenberghe will be a useful reference.

Grading policy

• No exams.

• Students will be evaluated by their homework assignments.

Lecture Notes

The course material is reproduced from the EE104: Introduction to machine learning by Sanjay Lall and Stephen Boyd at Stanford university, under their kind permission.

1. Course overview

2. Predictors

3. Validation

4. Features

5. Empirical risk minimization

6. Constant predictors

7. Non-quadratic losses

8. Non-quadratic regularizers

9. Neural networks

10. Classification

11. ERM for classification

12. Boolean classification

13. Multi-class classification

14. Probabilistic classification

15. ERM for probabilistic classification

16. Unsupervised learning

17. Principal components analysis

18. Optimization

19. Prox-gradient method

Assignments

Several sets of occasional homeworks will be assigned. You are encouraged to work in groups, however everyone should turn in his/her own work.

Files

These are some data and the Python codes in .ipynb notebook files that we are using for lectures or homework assignments.