Week 3, 5–9 October
Welcome to week 3! This week we introduce classification and gradient‑based fitting.
Many applications of machine learning use discrete labels (spam vs not spam) rather than real-valued outcomes (strength of metal alloy). We look at two different ways to “classify” (predict discrete labels), based on the main tools we’ve built up so far: regression and multivariate Gaussians.
This week also has the first discussion group meetings, and is the first week to be assessed. Your mark for this week depends on completing the discussion task (10%), attempting the in-note questions (20%), and this week’s assessed questions (70%). Full details on the assessments page including rules you must know.
Recommended event (entirely optional):
Distinguished lecture Monday 5 October, 3–4pm:
Christopher Bishop - Research and the “AI Revolution”
Here is what you need to do in Week 3:
- Any catch-up: Make sure you have done last week’s activities, especially the question sheet, which you will build on this week. Don’t get stuck on any “further reading” or optional parts, because you need to get on with this week. If there are any threads on hypothesis that didn’t get resolved (allow 48 hrs), email Arno and Iain.
- Lecture notes: Work through the Week 3 notes, answering all the questions. It’s fine to make mistakes here, but an honest attempt at these by Friday at 4pm (UK time) is required this week.
- Question sheet: Do the week 3 question sheet. This question sheet is assessed and forms the bulk of this week’s marks.
- Tutorial group discussions: We will be using Microsoft Teams for the tutorials. To get started with your MLPR tutorial discussion group, please read the MLPR tutorial discussion group instructions. This week’s discussion task is about the cancer classifier from the notes. In the Week 3 notes on Classification: Regression, Gaussians, and pre-processing, we considered the problem of creating a classifier for the Wisconsin Breast Cancer Data. In your tutorial group, create a list of things that you would consider if asked to create such a classifier, and questions you would ask the medical practitioners who are asking you to create the classifier. Your group discussion report on this task will be assessed! See the group instructions for details on how to submit the group discussion report.
We recommend that you aim to finish the questions (in the notes and question sheet) and submit your discussion report by the end of Thursday. We will assess only what you have submitted by 4pm UK time on Friday.
As in the Informatics late policy, extensions are not available for weekly hand-ins. We expect many students to miss or under-perform on one hand-in, and will discount the one with the lowest mark. If you experience more significant disruption to your studies, you may need to file special circumstances. Consult your Personal Tutor or Student Support team. Lecturers on a course cannot make allowances outside these procedures.