MA8701 General Statistical Methods

• We start on Tuesday January 8, and continue until Tuesday April 9, 2018.
• All weeks we will have lectures Tuesdays at 12.15-15.00, but week 8 (experimental design) we have lectures Tuesday 12.15-14.15 and Thursday 10.15-12.00
• All lectures will be in 734, 7. etg, sentralbygg 2, unless we find that this room is not meeting our needs.

Introduction to the course: week 2 08.01 (Thiago, Ben and Mette will present). Mette says a few words on statistical learning and the topics of this course, Thiago and Ben talk about their par. Thiago present the project work (based on a Kaggle data set) and how we plan for the project to be performed.

Part 1: week 3 15.01 and week 4 22.01: Statistical learning with sparsity (Benjamin Dunn)
Readings: Hastie et al (2015): Statistical learning with sparsity (selected chapters). We expand on what you have learned about the lasso and ridge in TMA4268, and marry with the GLM from TMA4315. See Reading list (left menu) for reading list and link.

More on project work: week 5 29.01: Erlend Aune will talk about how to perform computations for the compulsory project. Please bring your laptops, as there will be practical examples on how to schedule scripts on the GPU cluster. It is recommended to have gone through the information available at http://swcarpentry.github.io/shell-novice/ We will go through what is necessary to set-up a basic training loop on the cluster using Keras (python).

Part 2: week 6 05.02 and week 7 12.02: Smoothing and splines (Bo Lindqvist)
Readings: Friedman, Hastie and Tibshirani (2008): Elements of Statistical Learning. Chapter 5 and 6. Book at https://web.stanford.edu/~hastie/ElemStatLearn/ (free)

Part 3: week 8 19.02 and 21.02 Experimental design in statistical learning (John Tyssedal)
Readings: Book chapter and two articles - go to Reading list (left menu) for links.

Part 4: week 9 26.02, week 10 05.03 and week 11 12.03 : Deep neural nets (Thiago Martins)
Readings: François Chollet with J. J. Allaire (2018) Deep learning with R, https://www.manning.com/books/deep-learning-with-r or François Chollet (2017) Deep learning with Python https://www.manning.com/books/deep-learning-with-python. You choose what you read, both built on the keras package. This book you have to buy - at Manning as ebook or Akademika on paper. The library should also have 2 copies of the R book.

Part 5: week 12 19.03 and week 13 26.03 Active learning (Erlend Aune)
Readings: Review article, see Reading list.