Curricula

Objectives

You understand the significance of deep learning in the digitalized operational environment. You know about the most common methods of deep learning and how to apply them in practice to existing data as well as interpret the results of the methods.

EUR-ACE Competences:
Knowledge and Understanding
Engineering Practice
Research and information retrieval

Content

- Various neural networks and their architectures and uses (e.g. CNN, RNN, LSTM, Autoencoder).
- Work with open source neural network tools
- Transfer learning
- Prediction
- Machine vision
- NLP

Time and location

The course will be implemented in the spring semester of 2024.

Learning materials and recommended literature

The material for the assignments and the content to be studied will be shared during the course. The course utilizes the Python 3.9+ environment, Git version control, scikit-learn, TensorFlow 2, Keras and other visualization and deep learning libraries.

Teaching methods

Virtual study including doing assignments and familiarizing yourself with related lecture and example materials.

Practical training and working life connections

The aim is to connect the content of the course to problems that occur in working life.

Alternative completion methods

The admission procedures are described in the degree rule and the study guide. The teacher of the course will give you more information on possible specific course practices.

Student workload

The workload of one credit corresponds to 27 hours of study. The total amount of study work (5 ECTS) in the course is 135 hours.

Content scheduling

The learning materials are published at the start of the course and are supplemented if necessary during the course.
More detailed installation instructions for the environments used in the course are also given at the beginning of the course.

Further information for students

The course is evaluated using returned assignments.
The assessment methods are reviewed at the beginning of the course.

Evaluation scale

0-5

Evaluation criteria, satisfactory (1-2)

Satisfactory 2: You know the most commonly used techniques in Deep Learning for various problems. You know how to choose the techniques of Deep Learning and apply their technical know-how in practice. In addition, you know how to assess their implementation superficially.

Sufficient 1: You know the most commonly used techniques of Deep Learning. You are able to apply the most common techniques of Deep Learning. Additionally, you are able to assess their implementation briefly.

Evaluation criteria, good (3-4)

Very good 4: You recognize the advantages of Deep Learning in the era of digitalization. You know the most commonly used techniques of Deep Learning and is able to justify versatile the use of the implemented techniques in various tasks. You know how to apply their technical know-how in practice and assess their implementation critically as well as validate its development.

Good 3: You are aware of the advantages of Deep Learning in the era of digitalization. You know the most commonly used techniques of Deep Learning for various problems. You are able to apply his/her technical know-how in practice, assess their implementation in practice and validate its development.

Evaluation criteria, excellent (5)

Excellent 5: You recognize the advantages of Deep Learning in the era of digitalization. You know the most commonly used techniques and is able to critically justify the used techniques in various tasks. You are able to apply their technical know-how in practice and critically assess their implementation as well as validate its development.

Prerequisites

Basics in ICT, programming, knowledge and know-how of Python programming language.

Additionally, courses in Data Preprocessing and Computational algorithms.