Mathematics for Data Science

This course is over.

Lectures and Tutorials (6 SWS/9 ECTS)

This course provides a coherent treatment of the mathematical tools and techniques from stochastics, linear algebra and numerical mathematics underlying today's machine learning and big data technologies. Topics include probability theory, descriptive and inductive statistics, Markov chains, eigenspace analysis, matrix factorization, and sampling.

Knowledge equivalent to an introductory course on discrete mathematics is required for this class.

Times  

Lectures:Tuesday, 17:00-18:30, Z 1003 (Sven Kosub)
Thursday, 13:30-15:00, Z 1003 (Sven Kosub)
Tutorials:Wednesday, 15:15-16:45, Z 1003 (Julian Müller)
Oral exam:July 2016 (1st date)
October 2016 (2nd date)

Homework Assignments (local access only)

Content

The following topics are planned:

  • Descriptive statistics
  • Probability theory
  • Inductive statistics
  • Stochastic processes
  • Eigenspaces
  • Matrix factorization
  • Sampling

Lecture Notes (local access only)

German (!) lecture notes are made available close in time to the lectures. The current version can be downloaded here. In case you have suggestions or comments (typos or any kind of errors) please send an email.

Literature

In-depth and background material of certain course aspects can be found in (the list will be constantly updated):

  • Avrim Blum, John Hopcroft, Ravindran Kannan. Foundations of Data Science. An online textbook draft, 2016.
  • Peter Grindrod. Mathematical Underpinnings of Analytics. Oxford University Press, Oxford, 2014.
  • Ankur Moitra. Algorithmic Aspects of Machine Learning. An online textbook draft, 2014.
  • Olle Häggström. Finite Markov Chains and Algorithmic Applications. Cambridge University Press, Cambridge, 2001.
  • Thomas Schickinger, Angelika Steger. Diskrete Strukturen. Band 2: Wahrscheinlichkeitstheorie und Statistik. Springer-Verlag, Berlin, 2002.