TRAINING COURSE: Machine Learning - Fundamentals and Applications to Examples in Materials Science

Chair of the training course from April, 12th to 16th, 2021 is Dr.-Ing. Tim Dahmen, DFKI GmbH.


Artificial intelligence, specifically machine learning and deep learning is becoming increasingly important for the evaluation of materials science data, especially for image data.

In this training we offer you a practice oriented introduction to artificial neural networks for the automatic analysis of material science data. The focus will be on the classification and segmentation of image- and table- data.


Topics and contents

  • Deep Learning as a method of machine learning - basics and overview
  • From materials science to deep learning - examples and applications
  • Processing of tabular data with Artificial Neural Networks
  • Exercise I: Development of a Convolutional Neural Network (CNN) for the classification of table data
  • Processing of material science image data using convolution-based neural networks
  • Exercise II: Development of convolution-based neural networks (CNN) for the classification of image data
  • Exercise III: U-Net architectures for segmentation of material science image data
  • Manual and synthetic generation of training data
  • Summary


Your benefit

  • After a short introduction, which is not mathematically in-depth, application examples of Deep Learning are developed together.
  • You will learn how to implement and apply neural networks with the help of Python and suitable libraries. The focus is on the independent application of the developed models.
  • By executing and modifying the provided scripts on your own, you will be able to directly apply the acquired knowledge in practice.
  • After the participation you will know the possibilities and problems of machine learning, so that you can efficiently transfer and adapt the learned contents to your own data.


Target audience

Ideal prerequisites for successful participation in the training course are basic programming skills in Python, Matlab or other programming languages. The previous knowledge includes: variables and associated arithmetic operations, functions, case distinctions, control structures). Basic knowledge of mathematics is also helpful. For example, you should have an idea about the keywords vector, linear dependence, gradient and non-linearity.


More information on the contents and prices of this training can be found here.

If you have any questions for this training course, please contact us at


Sign up for our newsletter

Subscribe to our newsletter for regular updates about materials science topics!

After subscribing, you will receive an email from us with a confirmation link.
Only after clicking this link your registration is completed.

Terms and Conditions Privacy Policy Antitrust Notice Contact Imprint