Künstliche Intelligenz für Computational Mechanics

What it is about

You learn methods of machine learning and their application in numerical mechanics. The module covers fundamentals of supervised, unsupervised and reinforcement learning, deep neural networks, physics-informed networks, surrogate models and model-reduction techniques. By the end you will be able to implement ML techniques, evaluate them and apply them to problems of Computational Mechanics.

What you will be able to do

• implement simple ML algorithms independently
• apply ML techniques to technical problems
• evaluate and apply ML techniques in numerical mechanics

What the module consists of

• Lecture with integrated exercisesConvey theoretical concepts and directly implement them in exercises

Teaching method

• Lectures with integrated exercisesIntroduction to theory, followed by implementation in Python
• Practical implementation in Python with open-source frameworks and IDETo apply and verify the presented methods in practice
• Individual appointments and open Q&A sessions with the instructorSupport for exercises and projects