Physikbasiertes Machine Learning

What it is about

You will learn selected methods of machine learning from basic concepts to state of the art. The focus is on classification and regression, clustering and dimensionality reduction, as well as generative models; in the end you will be able to understand the methods, implement them, and apply them to real problems while considering physical boundary conditions and invariances.

What you will be able to do

• Understand key concepts of different ML algorithms
• Apply discussed methods to test problems
• Implement algorithms and transfer to real problem settings
• Integrate physical boundary conditions and invariances into ML methods
• Compare methods with regard to application, advantages/disadvantages, and limits

What the module consists of

• VorlesungVermittlung motivierender Beispiele, Schlüsselkonzepte und mathematischer Hintergründe
• ÜbungHands-on Erfahrung mit Machine Learning Techniken und praktische Anwendung

Teaching method

• VorlesungsfolienVorstellung motivierender Beispiele und Schlüsselkonzepte
• TafelanschriebErläuterung wichtiger mathematischer Hintergründe
• AnimationenDemonstration von Algorithmen
• Übungs-Handson mit PythonPraktische Umsetzung und Anwendung der Methoden (Lösungen in Python verfügbar)