Universität Wien FIND

270070 VO+UE Machine learning for molecules and materials (2020W)

with an introduction to python

4.00 ECTS (3.00 SWS), SPL 27 - Chemie
Continuous assessment of course work


Note: The time of your registration within the registration period has no effect on the allocation of places (no first come, first served).


max. 50 participants
Language: English



The lecture will be held online (until further notice) via BigBlueButton. Since group discussions are planned, there is a fixed lecture time: Thursdays, 10:00-13:00. During this time, also the computer exercises will take place. First date: October 8, 2020.


Aims, contents and method of the course

The basic concepts of machine learning in theoretical chemistry for molecular and materials systems will be presented. First, a general introduction of important machine learning models like neural networks and kernel-ridge regression will be provided. Python as programming language will be introduced for translating the learned concepts into computer codes. The concepts will be expanded and applied for molecular systems. Different types of representations of molecules will be discussed. Furthermore, machine learning for materials will be presented. The content will be presented in an intertwined sequence of presentations, sessions of flipped classroom and computer exercises.

The goals are:
- Understanding of and overview over machine learning methods for molecular and materials systems.
- Ability to write small programs in python, with a focus on machine learning for theoretical chemistry.
- Knowledge on solving problems with machine learning codes.

Assessment and permitted materials

Performance will be assessed through short tests and an oral presentation as well as the participation during the computer exercises and the lecture. It is possible to prove the performance by solving a problem in chemistry through machine learning.

Minimum requirements and assessment criteria

Basic knowledge of theoretical chemistry (Bachelor level: Hartree-Fock, harmonic oscillator, mathematical basics, etc.) is required. The grade consists of the average results of the short tests (30%), the oral presentation (30%) and the participation (40%).

Examination topics

Content of the lecture.

Reading list

- C. Bishop, Pattern recognition and machine Learning
- https://www.deeplearningbook.org/
Research articles as discussed during the oral presentations.

Association in the course directory

PC-4, D.4, MC-3, D.3, Doktorat

Last modified: Mo 05.10.2020 16:29