Department of Computer Science and Technology

Course pages 2018–19

Foundations of Data Science

Lecture notes and example sheets

  • Full notes [pdf]
    Sections marked * are non-examinable, but they may be helpful background reading for the example sheets. Material in the example sheets is examinable.
  • Example sheet 0 [pdf]: review of IA Maths for NST (not intended for supervision)
    solutions [pdf]
  • Example sheet 1 [pdf]: probability and random variables
    practical 1 [pdf]
  • Example sheet 2 [pdf]: inference
    hint sheet [pdf]
    mock exam questions [pdf]; solution 1 [pdf], solution 2, solution 3 [pdf]
    practical 2 [pdf]
  • Example sheet 3 [pdf]: model crafting
    hint sheet [pdf]
    mock exam questions [pdf]; solution 4, solution 5 with supervisors
    practical 3 [pdf] with hint sheet for q1 [pdf]

Errata

When I make mistakes during lectures, I upload the corrected slides to Moodle. For errors in the lecture notes, here is the list of errors discovered so far.

  • Section 1.2 page 3 exercise 1.1. The display equations for P(Xx) should have x outside the summation, not k. (Thanks to LAM.)
  • Section 1.3 page 5 example 1.6. The first bullet point should have Z=2, not Z=4.
  • Section 3.1 page 34 and appendix page 104. The formula for the normalizing constant for Beta is wrong.
  • Section 3.1 page 35. The integrals for the posterior predictive probability should be , not dx.
  • Lecture 5. In lectures, I wrote down the wrong formula for posterior mean. And I got the error probability wrong: 1-0.77 is equal to 23% not 28%.
  • Example sheet 2 question 2. Should say μ~Normal(μ002) rather than Normal(μ00)2.

Slides from lectures

— on Moodle


Timetable and announcements

— on Moodle


Code snippets

— posted one by one at notebooks.azure.com