Lecture notes

Sections marked * are non-examinable, but they may be helpful background reading if you want to study further in machine learning and data science.

• Notes parts I & II as handed out
• Notes part III section 6 as handed out
• Notes part III section 7 as handed out
• Notes part IV as handed out

There are also extended notes, incorporating all the material for this course and more. All this extra material is non-examinable, and is provided purely for interest.

• Extended notes DRAFT, editing in progress

Example sheets

These links include both example sheets (intended for supervision) and supplemental questions (intended for revision, unless your supervisor directs you otherwise). All tips and techniques described on the main example sheet are examinable.

• Example sheet 0 with solutions: review of IA Maths for NST (not intended for supervision)
• Example sheet 1: learning with probability models. (2019-10-20 23:00: updated with further hints)
• Example sheet 2: Bayesian inference
• Example sheet 3: frequentist inference
• Example sheet 4: Markov chains
• Practical 4 theory and code, also in Azure notebooks

Examples classes

Some lectures will be given over to worked examples, of mock exam questions. (The timetable for these lectures will be on Moodle.) These lectures are optional—some students find them helpful, other students find them a waste of time. You should attempt the question yourself in advance of the lecture, and decide whether or not to attend.

• Example class 1: learning with probability models
• Example class 2: Bayesian inference – model solution
• Example class 3: Frequentist inference – model solution
• Mock exam question 4: Markov chains – model solution