Prerequisites
Course prerequisites
Approximately 60% of the students taking the course lack some of the course prerequistes. For that reason, we have created a problem set that tests you in the primary material that we expect you to know beforehand or learn during the first few weeks of the course. The problem set is due two weeks into the course. If you can complete this problem set, you can assume you have the required prerequisites for the course. You can find the 2023 version of the problem set here.
The first week of the course consists of one lecture (45 minutes) that introduces the course (with no preparation needed), and then the remaining time is used for individual refreshing essential course prerequisites. TA’s will be present to help you with course prerequisites or problem set 1.
Overview of Prerequisites
| Prerequisites | Topics | Reading Material |
|---|---|---|
| Linear algebra | Matrix algebra, linear projections, PCA/SVD | ML 19.1–19.3, A.1 (online)–A.2 |
| Fourier series | Decomposition of signals | DSP 2.1–2.2, 2.6–2.7, 9.1–9.1.6 |
| Linear systems and Signal | LTI, Fourier transform, convolutions, sampling, quantization | DSP 1.1–1.7, 1.9, 3.1–3.3, 3.5–3.6, 5.1–5.3.1 |
| Probability theory | Probability density functions, Bayes theorem, expectations | ML 2.1–2.3 |
| Machine learning | Cross-validation, Linear regression, K-nearest neighbor, K-means | ML 1.1–1.5, 3.4, 3.12–3.13, 7.1–7.5, 12.6.1 |
Additionally, we have written a DSP primer that summarized (and a bit beyond) the DSP toolkit, including both Matlab and python examples.
The repository is located here: https://github.com/philgzl/dsp-primer
And a nicely compiled notebook is located here: https://nbviewer.jupyter.org/github/philgzl/dsp-primer/blob/master/notebook.ipynb
If you have not taken a digital signal processing course before, I strongly recommend you spend the first week working through the primer thoroughly.
Extra material
I have also compiled a list of resources for those who are missing some of the course prerequisites. These are full play-lists of courses so please do your own cherry-picking of what you need further material on: