Machine Learning 2 (Summer Term 2026)

Overview

• Course (2/2/0) consisting of:
  • Lectures in APB/E023/U on Mondays, 11:10–12:40
  • Exercises in APB/E023/U on Tuesdays, 11:10–12:40, starting 2026-04-21
  • Self-study
  • Final examination
• Lecturer: Bjoern Andres
• Teaching Assistants: Lucas Fabian Naumann, David Stein
• Enrolment (OPAL). Additional rules for enrolment may apply, depending on the study program.

Contents

This course is about selected, not necessarily connected topics of machine learning and its applications:

• Introduction
• Linear and integer optimization for machine learning
  • Simplex algorithm
  • Branch-and-bound algorithms
  • Branch-and-cut algorithms
  • Partial optimality
• Supervised learning
  • Basics (from Machine Learning 1)
  • Learning of rectified linear units
  • Learning of support vector machines
• Unsupervised learning
  • Basics (from Machine Learning 1)
  • Clustering
    • Problem statement, hardness
    • Linear program relaxations
    • Branch-and-cut algorithms
    • Partial optimality
    • Application: Image segmentation
  • Ordering/preordering
    • Problem statement, hardness
    • Linear program relaxations
    • Branch-and-cut algorithms
    • Partial optimality
    • Application: Social network analysis
• Supervised structured learning
  • Basics (from Machine Learning 1)
  • Graphical model inference
    • Problem statement, hardness
    • Linear program relaxations
    • Branch-and-cut algorithms
    • Partial optimality
    • Application: Pixel classification
• Embedding
  • Principal component analysis
  • Auto-encoders
  • Variational auto-encoders
  • Application: Image generation and interpolation
• Graph neural networks