Machine Learning for Molecular Physics WiSe 22/23

16 lectures & 15 tutorials

Lectures: Friday 16:00-18:00 - 1.1.26 Seminarraum E1 (Arnimallee 14) or ONLINE

Tutorials: Wednesday 16:00-18:00 - 1.4.03 Seminarraum T2 (Arnimallee 14)

First day of lectures: October 17, 2022

Last day of lectures: February 17, 2023

First tutorial: October 26, 2022

Winter break: December 17, 2022 – January 2, 2023

CHAT ROOM: 

https://app.element.io/#/room/!drONOPsNDRzWOsdXBa:physik.fu-berlin.de

You need an account in Physics to be able to join the chat. Please ask zedat to give you an account in Physics if you don't have one already.

Requirements:

• Basic stat mech
• Knowledge of computational methods for molecular simulations; MD, MC
• Basic knowledge in solid state physics
• Basic programming in python or related

Lecture content (Friday 16:00-18:00 in person or online):

1. 21.10.: general intro (in person)
2. 28.10.: Introduction to ML, general (in person)
3. 04.11.: Introduction to ML for Molecular Simulations, an overview of possible applications, comparing classical approaches (online)
4. 11.11.: ML for molecular simulation analysis: theory of Reaction Coordinates (online)
5. 18.11.: ML for molecular simulation analysis: ML for dimensionality reduction (in person)
6. 25.11.: ML for force fields/potential energy surface bulk+molecules - I (in person)
7. 02.12.: ML for force fields/potential energy surface bulk+molecules - II (in person)
8. 09.12.: ML for force fields/potential energy surface bulk+molecules - III (in person)
9. 16.12.: ML for force fields/potential energy surface bulk+molecules - IV (online)
10. 06.01.: ML for classification, structure, and properties: condensed matter (online)
11. 13.01.: ML for classification, structure, and properties: biophysics (online)
12. 20.01.: ML for classification, structure, and properties: molecules (online)
13. 27.01.: ML for learning free energy landscapes - I (online)
14. 03.02.: ML for learning free energy landscapes - II (online)
15. 10.02.: Boltzmann Generator and related methods (in person)
16. 17.02.: future perspectives and directions, what can ML do for Physics (in person)