Assessment

Grading and the dates of the 2020 offering in one place, as announced on the course page. The project carries the most weight; homework assignments are programming tasks, a subset of them with a viva.

Dates

Date Event
06 Jan 20 First lecture
15 Jan 20 Quiz 1 held by this date
20 Jan 20 Project proposal report due
10 Feb 20 Project phase I presentations (week of)
15 Mar 20 Project phase II presentations (week of)
30 Mar 20 ML demo due
20 Apr 20 Three-minute project pitch, demo and poster (week of)
22 Apr 20 Final project report due
13 Jul 20 Last lecture

Project and quiz dates are as announced at the start of the semester; lectures continued until July.

Grading

Everything is out of 100 marks, with bonus marks on top.

ComponentMarksHow
Project40In groups of four or fewer: proposal report, due 20 January (5); phase I presentation, week of 10 February (5); phase II presentation, week of 15 March (5); three-minute pitch in the style of 3MT, week of 20 April (5); demo and poster, week of 20 April (15); final report, due 22 April (5). Ideas: from the instructors and from a Stanford ML course.
Homework32Eight programming assignments. Late submissions earn 50% credit: up to one day late for the first assignment and up to two days for the others. A subset has an associated viva.
End semester10Written exam.
Quizzes6Three surprise quizzes worth 3 marks each; the best two count. Quiz 1 is held on or before 15 January.
Paper presentation4Three students per paper: one summarises it, one argues for it, one acts as devil's advocate, convincing the jury that the paper has flaws. Paper list · slides template.
Kaggle competition4Done individually, due in early April.
ML demo4An interactive demo, in the same team as the project, due 30 March. Examples: k-means, a visual introduction to ML, linear regression, random forests.
Total100Bonus: 10 marks for reaching Master level on Kaggle, or 5 marks for Expert level.