Weekly plans and update for week 36

Hi all, we hope the week started the best possible way. Here comes our weekly summary from last week and this week's plans (week 36).

Last week we ended with a discussion on statistics and probability theory, the lecture notes at https://compphysics.github.io/MachineLearning/doc/pub/Statistics/html/._Statistics-bs000.html give you some of the elements that were discussed, similarly chapters 2 and 7.1 of Murphy's text are also good reads, see https://github.com/CompPhysics/MachineLearning/blob/master/doc/Textbooks/MachineLearningMurphy.pdf

This week we start on Thursday with a repetition of some of the topics discussed last week and link this with the discussion of resampling techniques like Bootstrap and Cross Validation and the bias-variance trade-off. A good read is Hastie et al's text chapters 7.1-7.3 and 7.10 and 7.11. The lecture slides (towards the end) contain several code examples, see https://compphysics.github.io/MachineLearning/doc/pub/Regression/html/Regression-bs.html contain several code examples.

On Friday we will start discussing Ridge regression, again see the slides on https://compphysics.github.io/MachineLearning/doc/pub/Regression/html/Regression-bs.html

and Hastie et al chapter 3.4.

The exercise set 2 shows also some simple examples (with solutions) relevant for project 1.

We will also discuss project 1.

A quick reminder, the zoom link for the lectures is always available from the official schedule, similarly the zoom link for digital labs (labs 2 and 7), see https://www-adm.uio.no/studier/emner/matnat/fys/FYS-STK4155/h20/timeplan/index.html?vrtx=admin

New, if of interest, we will also, since we are several at the lab during labs 3-6, we would like to offer the possibility of access via zoom. Depending on how many may be present in person at the lab, response can be slow.

See you soon at the lab, where we discuss both project 1 and the exercise set for weeks 36 and 37.

Morten et al