Linear Algebra and Optimization for Machine Learning

State-of-the-art machine learning methods combine techniques from many different areas of mathematics. In this course, we will discuss algorithmic foundations from the areas of linear algebra and optimization. This also includes aspects of their theory and efficient implementation. Among others, we will study

• matrix decomposition and factorization techniques,

• regression and classification problems and regularization,

• iterative and continuous optimization methods,

• hyper parameter optimization,

• graph-based algorithms and clustering,

• basic introduction to neural networks and algorithmic aspects neural networks (e.g. computational graphs and back-propagation).

The final grade of the course consists of the following components:

• Two group projects (0.3)

  • Mandatory submission of project report

  • Assessment via group presentations followed by oral examination

• Written exam (0.7)

Final grade calculation: (0.3 * group projects + 0.7 * written exam)

Note: For the written exam, a score of at least 5.0 must be obtained.

Resit/ Repair opportunities:

In case of an insufficient result, repair opportunities may be offered in accordance with TER Implementation Regulations Art 5, sub 5., for:

• Group projects: resubmit deliverable

• Written exam: There will be a resit for the written exam at a later point in the same academic year

Disclaimer: information may change depending on unforeseen circumstances or measures (see: TER Art 29, sub 4).