Data Science
Lecture notes
— [datasci2026.pdf]. If you spot a mistake in these notes, please let me know.Announcements, timetable
— MoodleLecture schedule
This is the planned lecture schedule. It will be updated as and when actual lectures deviate from schedule. Material marked * is non-examinable. Slides are uploaded the night before a lecture, and re-uploaded after the lecture with annotations made during the lecture.
| Prerequisites | |
|
Example sheet 0 and solutions
NumPy tutorial |
|
| §1–§4. Learning with probability models | |
| Lecture 1 |
1. Learning with probability models
1.1 Specifying probability models
|
| Lecture 2 |
1.2 Standard random variables
1.3 Maximum likelihood estimation
1.4 Numerical optimization with scipy
|
| Lecture 3 |
1.5 Likelihood notation
1.6 Types of model
1.7 Supervised and unsupervised learning
|
| Lecture 4 |
3. Neural networks as probability models (* non-examinable)
|
| Lecture 5 |
2.1 Linear modelling
2.2 Feature design
|
| Lecture 6 |
2.3 Diagnosing a linear model
2.5 The geometry of linear models
|
| Lecture 7 |
2.6 Interpreting parameters
2.4 Probabilistic linear modelling
|
| Lecture 8 |
Discussion of climate dataset challenge (* non-examinable)
|
|
Example sheet 1
OPTIONAL ex1 practical exercises [ex1.ipynb] (for supervisions) OPTIONAL PyTorch introduction OPTIONAL climate modelling challenge
Datasets investigated:
climate.ipynb
stop-and-search.ipynb
|
|
| §5, §6, §8. Bayesian inference and Monte Carlo | |
| Lecture 8 ctd. |
4, 10. Generalization, model choice, and holdout sets
8. Bayesianism
5.1 Bayes's rule for random variables
|
| Lecture 9 |
8.1, 8.2 Bayesianism
5.2, 8.3 Bayes's rule calculations
8.4 Bayesian readouts
|
| Lecture 10 |
6.1 Monte Carlo integration
6.2 Bayes's rule via computation
|
| Lecture 11 |
8.7 Bayesian model building
|
|
Mock exam question 2
and walkthrough video (29:35)
OPTIONAL ex2 practical exercises
[ex2.ipynb] (for supervisions)
OPTIONAL Climate confidence challenge
Code snippets: bayes.ipynb
|
|
| §7, §9, §10. Frequentist inference and empirical distributions | |
| Lecture 11 ctd |
5.3 Deriving the likelihood
|
| Lecture 12 |
7.1–7.2 Empirical cdf
7.3 The empirical distribution
9. Frequentism
9.1, 9.2 Resampling / confidence intervals
9.5 Non-parametric resampling
|
| Lecture 13 |
9.3 Hypothesis testing
|
| Lecture 14 |
Approaches to generalization: summary
|
| §11+§14. Random systems | |
| Lecture 14 ctd |
11.1 Text as a random system
11.2 Causal diagrams
|
| Lecture 15 |
11.3 Markov chains, RNNs, transformers
11.4 Fitting a Markov model
RLHF (* non-examinable)
|
| Lecture 16 |
11.5 HMMs
12.1 Stationarity and average behaviour of Markov chains
|
| Example sheet 4 | |