tbd.XX My Hidden Text

This was run in 2014 - 2016, The material remains here for reference

In this book club we go beyond the classical statistical methods for modelling, such as linear regression. As computing power has increased over the last 20 years many new, highly computational, “Statistical Learning”, methods have been developed.

In particular the last decade has seen a significant expansion of the number of possible approaches. This book club will provide a very applied overview to such modern methods as Generalized Additive Models, Decision Trees, Boosting, Bagging and Support Vector Machines as well as more classical linear approaches such as Logistic Regression, Linear Discriminant Analysis, K-Means Clustering and Nearest Neighbors.

We will also introduce code and possibly discuss some of the exercises in the book.

The current plan is to meet bi-weekly starting on Thursday, Nov. 6th 2014 in ATC B11. We will try to cover half a chapter per session (except for chapter 2) and possibly add sessions on biological data analysis problems.

Please register here to receive the announcements (Internal access only)

About the book / slides / videos

We will discuss

Current Schedule

Date Time Room Chapter Topics
6.11.14 17:00 B11 Chapter 2 Intro to Stat. Learning, Prediction Accuracy vs Model Interpretability, Bias-Variance Trade-Off
2.12.14 17:00 A23 Chapter 3 I Linear Regression till 3.3.1 “Qualitative Predictors”
11.12.14 17:00 A23 Chapter 3 II Rest of chapter 3,
08.01.15 17:00 B11 Chapter 3 III R-lab on regression of chapter 3, Link to material
22.01.15 17:00 B11 Chapter 4 I Logistic Regression, Linear Discriminant analysis (up to 4.4.4)
05.02.15 16:00 A23 Chapter 4 II Logistic regression, microarray classification examples
19.02.15 16:00 B11 Chapter 5 I Cross Validation and Bootstrap, see Tim Hesterberg’s nice review on resampling
05.03.15 16:00 B11 Chapter 5 II Lab CV, feature selection using CV and caveats, Link to material, paper on selection bias
26.03.15 16:00 A23 Chapter 6 I Regularization: Subset regression, ridge and lasso penalties — paper on p–values for the lasso
16.04.15 16:00 A23 Chapter 6 II Lasso and feature selection: application to QTL mapping
07.05.15 16:00 A23 Chapter 7 I Splines and local regression
21.05.15 16:00 B11 Chapter 7 II Non–linear / smoothing methods: Lab with examples from RNA-Seq and HiC
09.02.16 16:15 B11 Chapter 8 / 9 Introduction to supervised learning with trees, random forests and SVMs. For additional material, see the caret package and accompanying book as well a recent evaluation of classifiers and of course DJ Hands classical paper on the illusion of progress in classifier technology.
18.02.16 16:00 B11 Chapter 8 / 9 Case studies on classification with metagenomics data with Georg Zeller, see e.g. this paper.