Grading and every dated event of the August–November 2022 offering in one place. The project carries the most weight; assignments are programming tasks done in pairs.
Grading
| Component | Weight | How |
|---|
| Project | 60% | See the project ideas. |
|---|
| Assignments | 20% | Four programming assignments, each a single Jupyter notebook done in a group of two. |
|---|
| Quizzes | 20% | Each quiz covers everything from the previous quiz up to the day of the quiz. |
|---|
| Total | 100% | |
|---|
Assignments
| Assignment | Topic | Marks | Due |
|---|
| Assignment 1 | Gradient descent with JAX; maximum likelihood for Normal, Student-t and exponential distributions | 5 | 10 Aug, 6:00 pm |
| Assignment 2 | MLE and MAP for linear and logistic regression; Normal, Laplace and Student-t choices | 6 | 23 Aug, 6:00 pm |
| Assignment 3 | Laplace approximation, Monte Carlo and probit approximation for logistic regression | 4 | 15 Sep, 6:00 pm |
| Assignment 4 | CNNs with MC dropout in Flax; Gaussian processes on the Mauna Loa CO2 data | 5 | 27 Oct, 6:00 pm |
The common rules are on the Assignments page.