MA8701 General Statistical Methods

Introduction and project work

• Introduction 08.01.2019: Presentation
• Github repository for introduction to parallell computing with GPUs on epic.hcp.ntnu.no (Erlend Aune)

Part 1: Regularized linear and generalized linear models

• Statistical learning with sparsity 15.01.2019: statistical_learning_with_sparsity_1.pdf
• Statistical learning with sparsity 22.01.2019: statistical_learning_with_sparsity_2.pdf
• Data for exercise: xdata.txt,ydata.txt

Part 2: Smoothing and splines

• Slides for Chapter 5 of Elements of Statistical Learning
• Slides for Chapter 6 of Elements of Statistical Learning
• Note on kernel density estimation

Part 3: Experimental design in statistical learning

• Note on response surface methods applied to a random forest example
• Note on nonparametric methods to compare algorithms on N data sets
• From TMA4267 Linear statistical methods: the DOE-note by John Tyssedal

Part 4: Deep learning

• 14 files (html and pdf) with slides/modulepage at https://www.math.ntnu.no/emner/MA8701/2019v/DeepLearning/
• Module page on neural networks from TMA4268 Statistical learning (not covering convnets and recurrent nets, but layered dense nets)

Part 5: Active learning

(press "download" to download the pdf-file)

• Slides Lecture 1: https://github.com/Froskekongen/MA8701/blob/master/lectures/MA8701_little_data_pdf_lecture_20190319.pdf
• Slides Lecture 2: https://github.com/Froskekongen/MA8701/blob/master/lectures/MA8701_-_Active_Learning.pdf