Slides

• T0 Course Organization
  
• T1 MachineLearning, UnivariateDataAnalysis v2
  
• T2 Algebra Theory, Lazy Learning
  
• T3 Model Evaluation
  
• T4 Probability Theory, Bayesian Learning
  
• T5 Information Theory, Feature Selection
  
• T6 Decision Trees, Patterns
  
• T7 Linear, Bayesian Regression
  
• T8 Perceptron, Gradient Descent v2
  
• T9 Neural Networks v2
  
• T10 Learning Theory, Model Selection
  
• T11 Dimensionality Reduction
• T12 Clustering
• T13 RBFs, SVMs, Kernel Machines

Disclaimer: the slides are offered as complementary reference material. Some of the covered contents are not necessary to know by heart, e.g. properties of different vector spaces (T2), ROC analysis and specifics of statistical testing (T3), coding theory (T5), derivation of bias-and-variance components (T10). More information on the contents to consolidate for the final exam to be released in the upcoming weeks.
Errata: delta1 in T9 updated in 17/10 (v2), Spearman calculus updated in T1 (v2), IQR-based outlier detection [Q1-1.5*IQR,Q3+1.5*IQR] in T1 slide 53

Exam preparation

• contents not covered in the exam: ROC analysis, coding theory, MDL, reinforcement learning
• first exam 2021/22
  
• exams from Machine Learning 2020/21
  • Exam 1 (solutions) - note: Ex.1.a t3=-2, Ex.3.a w=(-1,0.84,-1), Ex.4.a sum updated
    
  • Exam 2 (solutions) - note: Ex.2.a SUM_{k=1}^K should include all elements
    

Exam FAQ

• Please check the exam FAQ before contact your faculty hosts on exam-related matters.

• Is there a minimum grade for the exam?
  No. The minimum grade that was initially established for the exam is omissive in the version sealed by the Pedagogical Council, and due to the recency of this notification we were unable to grant a consensual reduction on the minimum grade for the exam. In this context, the professors of the course ask students to deligently prepare the contents for pedagogical reasons.