Fundamentals of Machine Learning [2021 Sommer]  

Code IFML 
Name Fundamentals of Machine Learning 

Leistungspunkte 8 LP 
Dauer one semester 
Turnus in (irregular) alternation with *Machine Learning* 
Lehrform 4 SWS lecture (in English), 2 SWS tutorial, homework assignments 
Arbeitsaufwand 240h; thereof 90h lectures and tutorials 90h lecture wrapup and homework 60h graded final report 
Verwendbarkeit cannot be combined with *Machine Learning* B.Sc. Angewandte Informatik, M.Sc. Angewandte Informatik, M.Sc. Scientific Computing 
Lernziel  Students understand fundamental concepts of machine learning (features vs. response, unsupervised vs. supervised training, regression vs. classification etc.), get to know established learning methods and algorithms, are able to apply them to realworld problems, and can objectively assess the quality of the results. In addition, students learn how to use Pythonbased machine learning software such as scikitlearn. 

Inhalt  The lecture, along with its sibling *Advanced Machine Learning*, offers an extended version of the onesemester course *Machine Learning*, with more room for regression methods, unsupervised learning and algorithmic details: Classification (nearest neighbor rules, linear and quadratic discriminant analysis, logistic regression, classical and randomized decision trees, support vector machines, ensemble methods); regression (linear and nonlinear least squares, regularized and sparse regression, robust regression); unsupervised learning (hierarchical clustering, kmeans algorithm, Gaussian mixture models and expectation maximization, principal component analysis, nonlinear dimension reduction); evaluation (risk minimization, model selection, crossvalidation) 

Voraussetzungen  recommended are: solid knowledge of basic calculus, statistics, and linear algebra  
Prüfungs modalitäten 
cannot be combined with *Machine Learning*, written exam (report on a 60h miniresearch project)  
Literatur  Trevor Hastie, Robert Tibshirani, Jerome Friedman: The Elements of Statistical Learning (2nd edition), Springer, 2009 