00EINEU016 20W 5SWS VI Computational Materials Design   Hilfe Logo

LV - Detailansicht

Wichtigste Meldungen anzeigenMeldungsfenster schließen
Allgemeine Angaben
Computational Materials Design 
lecture with integrated exercises
Winter semester 2020/21
Associate Professorship of Simulation of Nanosystems for Energy Conversion (Prof. Gagliardi)
(Contact information)
Allocations: 1 
Angaben zur Abhaltung
This course is the first step toward the paradigm shift of rational
materials design from purely theoretical methods based on physical
laws to a hybridation with automated learning strategies. Basic
quantum chemical theories would be introduced with a special focus on
Density Functional Theory. Then the theories would be applied to
predict some fundamental properties of materials using standard
software packages. Following this the basics of machine learning would
be introduced along with some hands-on applications. Machine learning
techniques would then be applied to predict material properties using
training data obtained by quantum mechanical calculations. The
potential of designing materials with desirably properties using a
combination of these two approaches would be explored.
The focus is on grad students (Open to Physics, Chemistry and Materials Science students also)
Basic programming
Helpful: Quantum mechanics, solid state physics
After successful completion of the module, students will

- understand the basics of quantum-chemical (QC) theories, with a special focus on Density-Functional-Theory (DFT)

- know which material properties can be predicted with QC/DFT-methods

-be able to use ab-initio software to model said material properties -be familiar with classical machine learning (ML) techniques and their theoretical foundations

- apply ML-techniques to simple datasets and evaluate the quality of the model

- gained insight into the current state of ML-based techniques for material property prediction

The course consists of weekly lectures and exercises. In the lecture the module contents will be presented by the teacher, supported by an electronic presentation. During the exercises, students will do hands-on calculations with numerical simulation programs/machine learning tools. In addition, there will be a final project.

e-Learning material on TUM-Moodle will be shared when the course is ongoing. ----
- presentation
- exercises solving computational problems
Für die Anmeldung zur Teilnahme müssen Sie sich in TUMonline als Studierende*r identifizieren.
Some small readings will be suggested during the course and will be announced in class

- Martin, R.M.: Electronic Structure. Basic Theory and Practical Methods. Cambridge University Press, 2004 [doi.org/10.1017/CBO9780511805769]

- James, G. et al.: An Introduction to Statistical Learning. Springer, 2013 [http://www-bcf.usc.edu/gareth/ISL

- Feliciano Giustino: Materials Modelling using Density Functional Theory
Online information
e-learning course (moodle)