remove chap 20 GP and re-add featsel as extra chapter

content/chapters/21_imbalanced_learning/21-01-intro.md → content/chapters/20_imbalanced_learning/20-01-intro.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 21.01: Introduction"
-weight: 21001
+title: "Chapter 20.01: Introduction"
+weight: 20001
 ---
 We define the phenomenon of imbalanced data sets and explain its consequences on accuarcy. Furthermore, we introduce some techniques for handling imbalanced data sets.
 <!--more-->

content/chapters/21_imbalanced_learning/21-02-perf-msr.md → content/chapters/20_imbalanced_learning/20-02-perf-msr.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 21.02: Performance Measures"
-weight: 21002
+title: "Chapter 20.02: Performance Measures"
+weight: 20002
 ---
 We introduce performance measures other than accuracy and explain their advantages over accuracy for imbalanced date. In addition we introduce extensions of these measures for multiclass settings.
 <!--more-->

content/chapters/21_imbalanced_learning/21-05-cs-3.md → content/chapters/20_imbalanced_learning/20-05-cs-3.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 21.05: Cost-Sensitive Learning 3"
-weight: 21005
+title: "Chapter 20.05: Cost-Sensitive Learning 3"
+weight: 20005
 ---
 We explain the concepts of instance specific costs and cost-sensitive OVO.
 <!--more-->

content/chapters/21_imbalanced_learning/21-06-cc-1.md → content/chapters/20_imbalanced_learning/20-06-cc-1.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 21.06: Cost Curves 1"
-weight: 21006
+title: "Chapter 20.06: Cost Curves 1"
+weight: 20006
 ---
 We introduce cost curves for misclassif error and explain the duality between ROC points and cost lines.
 <!--more-->

content/chapters/21_imbalanced_learning/21-07-cc-2.md → content/chapters/20_imbalanced_learning/20-07-cc-2.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 21.07: Cost Curves 2"
-weight: 21007
+title: "Chapter 20.07: Cost Curves 2"
+weight: 20007
 ---
 We explain cost curves with cost matrices and comparing classifiers. In addition we do a wrap-up comparision to ROC.
 <!--more-->

content/chapters/21_imbalanced_learning/21-08-smpl-1.md → content/chapters/20_imbalanced_learning/20-08-smpl-1.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 21.08: Sampling Methods 1"
-weight: 21008
+title: "Chapter 20.08: Sampling Methods 1"
+weight: 20008
 ---
 We introduce the idea of sampling methods for dealing with imbalanced data. In addition, we explain certain undersampling techniques. 
 <!--more-->

content/chapters/21_imbalanced_learning/_index.md → content/chapters/20_imbalanced_learning/_index.md

@@ -1,4 +1,4 @@
 ---
-title: "Chapter 21: Imbalanced Learning"
+title: "Chapter 20: Imbalanced Learning"
 ---
 This chapter introduces techniques for learning on imbalanced datasets. 

content/chapters/22_multitarget_learning/22-01-intro.md → content/chapters/21_multitarget_learning/21-01-intro.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 22.01: Introduction"
-weight: 22001
+title: "Chapter 21.01: Introduction"
+weight: 21001
 ---
 In this chapter we emphasize the practical relevance of  multi-target prediction problems. In addition, we name some special cases of multi-target prediction and establish the differences between transductive and inductive learning problems. 
 <!--more-->

content/chapters/22_multitarget_learning/22-02-losses.md → content/chapters/21_multitarget_learning/21-02-losses.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 22.02: Loss functions"
-weight: 22002
+title: "Chapter 21.02: Loss functions"
+weight: 21002
 ---
 In this chapter we introduce loss functions for multi-target prediction problems, explain the differences between instance-wise and decomposable losses and introduce the risk minimizer for both the hamming and 0/1 subset losses.  
 <!--more-->

content/chapters/22_multitarget_learning/22-03-methods-1.md → content/chapters/21_multitarget_learning/21-03-methods-1.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 22.03: Methods for Multi-target Prediction 1"
-weight: 22003
+title: "Chapter 21.03: Methods for Multi-target Prediction 1"
+weight: 21003
 ---
 In this chapter we introduce the concepts of independent models for targets, mean regularization, stacking and weight sharing in DL. 
 <!--more-->

content/chapters/22_multitarget_learning/22-04-methods-2.md → content/chapters/21_multitarget_learning/21-04-methods-2.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 22.04: Methods for Multi-target Prediction 2"
-weight: 22004
+title: "Chapter 21.04: Methods for Multi-target Prediction 2"
+weight: 21004
 ---
 In this chapter we introduce the Kronecker kernel ridge regression, graph relations in targets, probabilistic classifier chains and low-rank approximations. 
 <!--more-->

content/chapters/22_multitarget_learning/_index.md → content/chapters/21_multitarget_learning/_index.md

@@ -1,4 +1,4 @@
 ---
-title: "Chapter 22: Multitarget Learning"
+title: "Chapter 21: Multitarget Learning"
 ---
 This chapter introduces multitarget learning techniques. 

content/chapters/23_online_learning/23-01-intro.md → content/chapters/22_online_learning/22-01-intro.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.01: Introduction"
-weight: 23001
+title: "Chapter 22.01: Introduction"
+weight: 22001
 ---
 In this chapter we explain the differences between online and batch learning, the extended learning protocol in online learning and the strategies to measure performance in online learning.
 <!--more-->

content/chapters/23_online_learning/23-02-simple.md → content/chapters/22_online_learning/22-02-simple.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.02: Simple Online Learning Algorithm"
-weight: 23002
+title: "Chapter 22.02: Simple Online Learning Algorithm"
+weight: 22002
 ---
 In this chapter we introduce the formalization of online learning algorithms and the FTL algorithm. 
 <!--more-->

content/chapters/23_online_learning/23-03-ftl.md → content/chapters/22_online_learning/22-03-ftl.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.03: Follow the Leader on OLO problems"
-weight: 23003
+title: "Chapter 22.03: Follow the Leader on OLO problems"
+weight: 22003
 ---
 In this chapter we introduce OLO problems and explain why some FTL might fail on these problems. 
 <!--more-->

content/chapters/23_online_learning/23-04-ftrl.md → content/chapters/22_online_learning/22-04-ftrl.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.04: Follow the regularized Leader"
-weight: 23004
+title: "Chapter 22.04: Follow the regularized Leader"
+weight: 22004
 ---
 In this chapter we introduce FTLR as a stable alternative to FTL. 
 <!--more-->

content/chapters/23_online_learning/23-05-ftl-oqo.md → content/chapters/22_online_learning/22-05-ftl-oqo.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.05: Follow the Leader on OQO problems"
-weight: 23005
+title: "Chapter 22.05: Follow the Leader on OQO problems"
+weight: 22005
 ---
 In this chapter we prove that FTL works for online quadratic problems. 
 <!--more-->

content/chapters/23_online_learning/23-06-oco-1.md → content/chapters/22_online_learning/22-06-oco-1.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.06: Online Convex optimization 1"
-weight: 23006
+title: "Chapter 22.06: Online Convex optimization 1"
+weight: 22006
 ---
 In this chapter we introduce the class of online convex optimization problems and derive the online gradient descent as a suitable learning algorithm for such cases. 
 <!--more-->

content/chapters/23_online_learning/23-07-oco-2.md → content/chapters/22_online_learning/22-07-oco-2.md

@@ -1,6 +1,6 @@
 ---
-title: "Chapter 23.07: Online Convex optimization 2"
-weight: 23007
+title: "Chapter 22.07: Online Convex optimization 2"
+weight: 22007
 ---
 In this chapter we explain the connection between OGD and FTRL via linearization of convex functions and how this implies regret bounds for OGD.  
 <!--more-->

content/chapters/23_online_learning/_index.md → content/chapters/22_online_learning/_index.md

@@ -1,4 +1,4 @@
 ---
-title: "Chapter 23: Online Learning"
+title: "Chapter 22: Online Learning"
 ---
 This chapter introduces online learning. 

content/chapters/30_feature_selection/30-01-introduction.md

@@ -0,0 +1,14 @@
+---
+title: "Chapter 30.01: Introduction"
+weight: 30001
+---
+We motivate feature selection and discuss the difference to feature extraction.
+
+<!--more-->
+### Lecture video
+
+{{< video id="xiVB1EmlU9A" >}}
+
+### Lecture slides
+
+{{< pdfjs file="https://github.com/slds-lmu/lecture_sl/raw/main/slides-pdf/slides-fs-introduction.pdf" >}}

content/chapters/30_feature_selection/30-02-motivating-examples.md

@@ -0,0 +1,15 @@
+---
+title: "Chapter 30.02: Motivating Examples"
+weight: 30002
+---
+In this section, we explain the practical importance of feature selection and show that models with
+integrated selection do not always work.
+
+<!--more-->
+### Lecture video
+
+{{< video id="1BwgTptjDs4" >}}
+
+### Lecture slides
+
+{{< pdfjs file="https://github.com/slds-lmu/lecture_sl/raw/main/slides-pdf/slides-fs-motivating-examples.pdf" >}}

content/chapters/30_feature_selection/30-03-filters1.md

@@ -0,0 +1,15 @@
+---
+title: "Chapter 30.03: Filter Methods I"
+weight: 30003
+---
+We introduce how filter methods work and how they can be used for feature selection.
+
+<!--more-->
+
+### Lecture video
+
+{{< video id="RcDyvExpCSg" >}}
+
+### Lecture slides
+
+{{< pdfjs file="https://github.com/slds-lmu/lecture_sl/raw/main/slides-pdf/slides-fs-filters1.pdf" >}}

content/chapters/30_feature_selection/30-04-filters2.md

@@ -0,0 +1,15 @@
+---
+title: "Chapter 30.04: Filter Methods II (Examples and Caveats)"
+weight: 30004
+---
+In this section, we discuss how filter methods can be misleading and show how they work in practical applications. 
+
+<!--more-->
+
+### Lecture video
+
+{{< video id="X3FpzGnGA7o" >}}
+
+### Lecture slides
+
+{{< pdfjs file="https://github.com/slds-lmu/lecture_sl/raw/main/slides-pdf/slides-fs-filters2.pdf" >}}

content/chapters/30_feature_selection/30-05-wrapper.md

@@ -0,0 +1,15 @@
+---
+title: "Chapter 30.05: Wrapper Methods"
+weight: 30005
+---
+This section explains wrapper methods and explains how they can aid feature selection.
+
+<!--more-->
+
+### Lecture video
+
+{{< video id="XmvlHUCGNbc" >}}
+
+### Lecture slides
+
+{{< pdfjs file="https://github.com/slds-lmu/lecture_sl/raw/main/slides-pdf/slides-fs-wrapper.pdf" >}}

content/chapters/30_feature_selection/_index.md

@@ -0,0 +1,5 @@
+---
+title: "Extra Chapter: Feature Selection"
+---
+This chapter introduces feature selection, i.e., dinding a well-performing, hopefully small set of
+features for a task.