SDL Materials Design presents
Machine Learning for Materials Science
Date: October 21 - 23, 2024, 9.00 am - 01.00 pm; Format: online
Short abstract:
Data Science and Machine Learning are seen as the “Forth Paradigm” in Materials Science and are reshaping the research direction in many areas. In this training, the students/participants will gain an overview and obtain hand-on experience on the most relevant machine learning algorithms for theoretical simulations, experimental characterization, and in general statistical analysis in materials science. The participants will work with established packages within Python to develop their own simple machine learning based programs, and are going to tackle a challenging project. Though exemplary datasets, the participants will practice to apply appropriate methods to basic materials science problems, in particular Machine Learning assisted image segmentation, Machine-learning interatomic potentials, microstructure-property correlation analysis, and data-driven multiscale modeling.
Picture source.
Language: English
Capacity: 300
Requirement: On day 2 and 3, own laptop with working Google Collab account/ Jupyternotebook installed. A python installation with tensorflow, scikit-learn, ase, dscribe, and pacemaker
Further information: This training is organized in cooperation with the CRC FLAIR 1548 FLAIR – Start – FLAIR – TU Darmstadt and GRK 2561 MatCom-ComMat IAM – WerkstoffkundeKooperationen – GRK 2561.
Agenda
October 21, 2024, Day 1, 9.00 am – 1.00 pm
9:00 Motivation and examples of machine learning (ML) in material science (MS)
10:00 Break
10:15 Fundamental of machine learning, shallow learning vs. deep learning (DL),
convolutional neural network (CNN)
11:15 Break
11:30 Gaussian process (GP), Bayesian optimization (BO), microstructure descriptors
12:45 Q&A and feedback round
October 23, 2024, Day 3, 9.00 am – 1.00 pm
09:00 Motivation: Why do we need machine-learning in atomic simulations?
09:15 Introduction to atomic descriptors for machine learning on the atomic level
10:00 Break
10:15 Hands-on Tutorial: Make your own machine-learning model for NMR Shifts
11:00 Break
11:15 Machine Learning Potential: How to generate a database?
11:30 The Atomic Cluster Expansion: How to fit a potential?
12:45 Short feedback round
October 22, 2024, Day 2, 9.00 am – 1.00 pm
9:00 Introduction to Google Colab
9:15 Short refresher on Python and Introduction to PyTorch
9:50 Break
10:00 Fitting the Sine function with a polynomial: by NumPy arrays, PyTorch tensors,
Neural Network (NN) and Autograd from PyTorch
11:00 Break
11:10 Convolutional Neural Network (CNN) for predicting thermal conductivity of composites
12:00 Break
12:10 Gaussian Process (GP) and Bayesian Optimization (BO)
12:55 Q&A and short feedback round
Contact person
Aditi Mohadarkar
TU Darmstadt
Vasilios Karanikolas
TU Darmstadt
Prof. Dr. Bai-Xiang Xu
TU Darmstadt
Prof. Dr. Karsten Albe
TU Darmstadt