1. Course topics:
·
Review of selected
Calculus III topics (2 - 3 class meetings)
·
Review of selected
Linear Algebra topics (3 - 4 meetings)
·
Essentials of
probability and statistics (3 - 4 meetings)
·
Machine learning
landscape (3 meetings)
·
Regression: linear,
polynomial, logistic; multiple regression (4 meetings)
·
Dimensionality
reduction and feature extraction (5 meetings)
·
Neural networks (8 meetings)
·
Support Vector
Machines (4 meetings)
·
Naive Bayes classification (if time allows)
·
Clustering in
unsupervised machine learning (if time allows)
·
K-means algorithm
(if time allows)
The tentative schedule of the course has been posted on my website; see https://home.sandiego.edu/~pruski/m494s24schedule.html .
2. Course learning outcomes:
Upon successful completion of this course, the student will:
· Demonstrate a working knowledge of the mathematical foundations of machine learning, in the areas of calculus, linear algebra, and probability and statistics
· Demonstrate a working knowledge of such basic methods of machine learning as regression, dimensionality reduction, neural networks, support vector machines
· Be able to:
· apply methods of calculus, such as gradient descent, least squares minimization, chain rule for functions of several variables
· apply methods of linear algebra, such as eigendecomposition and singular value decomposition
· randomly generate synthetic data with required probability distributions
· perform and explain such methods of machine learning as various types of regression, PCA, neural network-based learning
· write code that implements the basic methods of machine learning
· communicate mathematical ideas clearly.
3.
Course
challenges
The three main challenges facing the course are:
· This is the first time a course of this type is taught as a regular class; although I taught the course four times, as MATH 499 or COMP 499, the dynamic of working with an individual student is completely different than teaching the course to a large class. Consequently, many things may not go as planned, in particular, we may not be able to exactly follow the schedule listed above.
· There is no textbook for the course; I have about 15 machine learning books and, for various reasons, none is suitable as the required reading in the course. I officially recommended one book to the Bookstore, and I will recommend four or five others in class. I will recommend reading Web-based materials, but be aware that so many of them are sloppy and contain errors, not only in spelling or grammar.
· This is a math course, but it has a strong programming component. You will have to do quite a lot of simple programming in Python. However, I will do my best to accommodate those of you who have had only a minimum exposure to programming.
4. Regular attendance is strongly recommended!
5. Office hours (Dr. Lukasz Pruski, Serra 147):
|
Monday |
12:00
- 2:00 |
|
Wednesday |
12:00
- 1:00 |
|
Thursday |
3:00 - 4:00 |
|
Friday |
12:00
- 1:00 |
and at other times, by appointment. (NOTE: These days/times are tentative at this point.)
6. Contact: The best way to contact me is by using e-mail (pruski@sandiego.edu or lukaszpruski@gmail.com). I read email many times per day. If for some reason you are unable to contact me, try calling our departmental Executive Assistant, Andrea, at extension 4706.
7. Course website can be found at https://home.sandiego.edu/~pruski/m494s24.html .
8. Homework assignments will be assigned and collected weekly. The assignments will usually involve writing code in Python. The total homework assignment score will count for 20% of the course grade.
9. A small research project (in teams of 2-3 students) will be assigned in early April to be completed by May 6. The project topics will include the machine learning material not covered in class. Each project requires a write-up, a program, and a class presentation. The project counts for 10% of the course grade.
10. There will be weekly quizzes, where quiz questions will refer to the recently covered material as well as to material you are supposed to read. We will determine the quiz schedule in class. Two lowest quiz scores will be dropped, and the remaining scores will count for 25% of the course grade. Quizzes cannot be made up unless you have a valid reason for not taking the quiz and you notify me in advance of your absence.
11. The midterm exam will take place on Friday, March 15. The test score will count for 15% of the course grade. A test can be made up only if you have an actual emergency and if you notify me in advance about your absence.
12. The final exam (Monday, May 20, 2:00) will be cumulative and its score will count for 30% of the course grade.
13. Grading criteria are as follows:
|
Total percentage |
Grade |
|
90% and above |
A |
|
80% - 90% |
B |
|
60% - 80% |
C |
|
50% - 60% |
D |
|
below 50% |
F |
Note: I will “curve grades up” in the unlikely case that the number of A’s and B’s falls below about 40% of the current enrollment.
17. The Mathematics Department strongly promotes Academic Integrity. I hope issues related to academic integrity will not arise in our course. There have been some cases of cheating in math courses in the past – mainly the cases of submitting someone else’s work as well as cases of cheating during exams. Depending on the severity of the case, the possible consequences include: assigning the score of 0 on the given assignment, lowering the course grade, or even assigning an F in the course. The USD academic integrity policy can be found at https://www.sandiego.edu/conduct/documents/Honor-Code.pdf).
18. Accommodations: Any student with a documented disability needing academic adjustments or accommodations is requested to speak with me during the first two weeks of class. All discussions will remain confidential. A student attempting to access Disability Services for the first time should begin by contacting the Disability and Learning Difference Resource Center (DLDRC) in SH, Room 300 (619/260-4655), e-mail: disabilityservices@sandiego.edu , website: www.sandiego.edu/disability/ It is the student's responsibility to schedule an "intake" meeting with the DLDRC Director as soon as possible.
19. Health Resources: The pandemic
has taught us that we need to change the way we behave. If you feel sick, please stay home to keep others healthy. The
following USD resources are available to students:
·
· MyWellness Portal: https://mywellness.sandiego.edu/