### Lectures

**13.09.2017. Lecture 1 - The Introduction**In the first lecture, you will hear about how machine learning and data science become more and more relevant in every aspect of human life. We then detail the course logistics and if time permits start to describe the formal setup of a machine learning problem.

Machine Learning at 720 degrees

doing good machine learning is like making good wine

first 24 minutes of this talk by Prof. Tom Mitchell (CMU) give a good intro

**15.09.2017. Lecture 2 - The Elements of Machine Learning**In this lecture, you will hear about the main components of a machine learning problem and its solution. We will discuss how to translate raw data into features and labels. One of the main goals of machine learning is to find a useful mapping from the features to labels. Such mappings are known as predictors (for continuous-valued labels) or classifiers (for discrete-valued labels). We will discuss the notion of loss and empirical risk for assessing the quality of a particular predictor or classifier.

lecture slides(with annotations)

Machine Learning for Optimized Eating (Miils Startup)

lecture of Prof. Tom Mitchell illustrating the elements of ML in the context of a particular ML method called ``decision trees''

lecture of Prof. Tom Mitchell on the basic principles of probabilistic machine learning

**19.09.2017. Lecture 3 - Regression I**In this lecture, you will learn some basic methods for solving regression problems where we want to predict a continuous label or target from features of a data point.

lecture slides (fixed typos, sep 22)

video lecture on regression by Prof. Andrew Ng (Stanford)

**22.09.2017. Lecture 4 - Regression II**In this lecture, we will already start covering some powerful regression methods based on the basic linear regression formalism. We will extend linear regression to full-blown non-linear models by kernel regression. We will also consider probabilistic learning setting and discuss Bayesian linear regression, where we can quantify the uncertainty in the model. Both approaches are widely used in many domains.

lecture slides**26.09.2017. Lecture 5 - Classification I**In this lecture, you will learn two of the most popular methods for classification, i.e., logistic regression and support vector classifiers (machines).

slides for Impromptu on Bayesian linear regression (with annotation)

lecture slides (with annotation)

Question and Answers for Lecture 5

video lecture on classification by Prof. Andrew Ng (Stanford)

lecture notes of Prof. Tom Mitchell (CMU) on logistic regression

lecture notes of Prof. Andrew Ng (Stanford) on support vector classifier (machine)

**29.09.2017. Lecture 6 - Classification II**In this lecture, you will hear about one non-parametric classification method, i.e., decision trees and one generative classification method, i.e., Bayes classifier.

lecture slides (annotated slides) additional notes on classification

Chapter 9.2 of this book contains more details on decision trees

notes on Gaussian random vectors

lecture notes by Prof. Tom Mitchell on (naive) Bayes classifier

paper on discriminative vs. generative method (for those who want to dig deeper)

video lecture on (naive) Bayes classifiers

**Lecture 7 - Validation**In this lecture, you will hear about one of the central challenges within ML, i.e., how to validate a particular ML method. By validating an ML method or model, we can assess how reliable the obtained predictions are. It is typically very helpful to have not only a single quantitative prediction but also a measure of how confident we should be in this prediction.

lecture slides (annotated slides)

Paper on Bias, Variance and the Combination of Least Squares Estimators.

The Bootstrap paper. The bootstrap is a method which allows to generate new synthetic datasets (e.g, a test set) by random resampling from the original dataset.

**Lecture 8 - Model Selection****Lecture 9 - Clustering**lecture slides; (with annotations) additional notes

video lecture of Prof. Ng on clustering as one of the fundamental unsupervised learning methods

video lecture of Prof. Ng on the basic formulation of the k-means clustering algorithm

video lecture of Prof. Ng on the optimization problem underlying k-means

video lecture of Prof. Ng on how to initialize k-means

video lecture of Prof. Ng on how to choose the parameter "k" for k-means

Paper on convergence of k-means algorithm

**Lecture 10 - Feature Learning**guest talk by Prof. Asokan on machine learning for internet security

lecture notes of Prof. Ng on principal component analysis (PCA)

video lecture of Prof. Ng on feature learning as dimensionality reduction or data compression

video lecture of Prof. Ng on PCA

**Lecture 12 - Recap**