Machine Learning Course

Within the framework of KTUR, the partners have consolidated their competencies in the field of continuing education and propose application-oriented continuing education courses based on the current needs of the companies in the border region.

Machine Learning Course

Explore the world of Machine Learning and put the methods and concepts you have learned directly into use with practical exercises practical exercises on real-world data.

The aim of the introductory course is to familiarize yourself with the topics of Machine Learning and Artificial Intelligence. You will acquire the theoretical basics and apply them directly through practical exercises on real data.You will learn how to process data and classical algorithms. We will use
Python, Scikit-learn and Kaggle.

The aim of the advanced course is to develop an understanding of deep learning and data visualisation. You will gain theoretical knowledge of the different components and architectures of neural networks and apply it to real-world
data via supervised and unsupervised approaches. We will use Python and Tensorflow.

Target groups & prerequisites

Target group:
Engineers, statisticians, computer scientists, physicists, and anyone with a mathematical/technical background.

Prerequisites:
First programming experience

Target group:
Engineers, statisticians, computer scientists, physicists and all with mathematical/technical background.

Prerequisites:
Programming experience with Python and initial experience using machine learning algorithms.

Program

 Morning, 9:00 – 12:00Afternoon, 13:00 – 16:00
Wednesday, June 8, 22T: Introduction to Artificial intelligence
PW: Data understanding with small datasets
T: Regression algorithms
PW: Implementation of one-dimensional and multidimensional
regression algorithms. 
Thursday, June 9, 22T: Classification algorithms
PW: Prediction of the semiconductor production yield
T: Clustering algorithms
PW: Evaluation of clustering algorithms
Friday,
June 10, 22
T: Time series analysis
PW: Analysis of Covid19 infection rates
T: Neural Networks: Multilayer perceptron
PW: Character recognition with neural networks

T: Theory – PW: Practical Work

 Morning, 9:00 – 12:00Afternoon, 13:00 – 16:00
Wednesday, June 22, 22T: Introduction to Deep Learning, Convolutional Neural Networks
PW: Segmentation and classification
T: Architectures and cost functions
PW: Regression and classification
Thursday, June 23, 22T: Advanced training: augmentation and dropout
PW: Segmentation with augmentation
T: Transfer learning, pre-trained architectures
PW: Transfer Learning with Deep Neural
Friday,
June 24, 
22
T: Dimension reduction and visualisation
PW: Eigenfaces
T: Stacked, sparse and denoising autoencoders
PW: Representation learning

T: Theory – PW: Practical Work

Organizational details

  • Introductory course: June 8-10, 22, 9:00-12:00 and 13:00-16:00 respectively
  • Advanced course: June 22-24, 22, 9:00-12:00 and 13:00-16:00 respectively
  • Introductory course: Karlsruhe, Germany*
  • Advanced course: Strasbourg, France*

* If the current COVID-19 restrictions do not allow in-person events, the courses will take place online.

The seminar will be held in English.

  • Single course (3 days respectively): 1.300 € 
  • Entire training course (6 days): 2.500 €

Please download the registration form and send it filled out to Ms. Romina Junk: romina.junk@h-ka.de.

Lecturer profiles

Professor at the Karlsruhe University of Applied Sciences.
Lectures in the Bachelor and Master programs at the Faculty of Electrical
Engineering and Information Technology.

Fields of expertise:

  • Systems theory
  • Signal Processing
  • Design For Six Sigma

Work Experience:

Researcher at Forschungszentrum Karlsruhe, developer and Product Owner at Robert Bosch GmbH. Developer and trainer of statistical methods of Design for Six Sigma.

Chair of Data Science and Artificial Intelligence at Télécom Physique Strasbourg and the ICube research laboratory, University of Strasbourg.

Fields of expertise:

  • Deep Learning
  •  Representation learning and clustering
  • Unsupervised approaches
  • Domain adaptation
  • Medical imaging and remote sensing.


Work experience:

Alumnus of the University of York and the U.S. Department of State’s International Visitor Leadership Program. Different positions in industry and academia, among others with QinetiQ Ltd. and the UK Ministry of Defence

Contact

If you have any questions, please contact:

Romina Junk
Hochschule Karlsruhe
romina.junk@h-ka.de
0721 925 2800

Copyright Picture: istock.com/PeopleImages