Machine Learning in Materials Processing & Characterization

Course curriculum and materials for a 4th semester undergraduate course (5 ECTS) focusing on applying machine learning techniques to experimental materials science data.

Course Overview

This course provides students with essential skills and practical knowledge to harness machine learning techniques for accelerating materials discovery and design. Specifically tailored for students interested in the BSc program "KI-Materialtechnologie" (AI for materials technology), it provides hands-on experience with core and advanced machine learning methods—including neural networks, optimization strategies, and generative modelling—to tackle real-world materials science problems.

Key Focus Areas:

• Machine learning techniques for materials processing and characterization data
• Vision-based ML for microstructure analysis and classification
• Time-series ML for process monitoring and optimization
• Spectral data analysis using dimensionality reduction and ML
• Multi-modal data fusion combining images, spectra, and process parameters

Course Structure

Duration: 14 weeks
Credits: 5 ECTS
Format: 2h lecture + 2h exercises per week

The course is organized into five units:

1. Foundations: From Materials Signals to Machine Learning (Weeks 1-3)
2. ML for Microstructure: Vision & Representation (Weeks 4-6)
3. ML in Processing: Time-Series, Optimization, Thermal/Mechanical Data (Weeks 7-9)
4. ML for Characterization Signals (Weeks 10-12)
5. Project + Reflection (Weeks 13-14)

Prerequisites

• Basic knowledge of Python programming
• Familiarity with machine learning fundamentals (covered in parallel ML intro course)
• Understanding of materials science fundamentals

Learning Outcomes

Upon completion of this course, students should be able to:

• Interpret materials characterization and processing data in an ML-ready way
• Build ML pipelines for microstructure classification, process prediction, and spectral analysis
• Understand the physics of image/signal formation well enough to avoid "garbage in → garbage out"
• Evaluate uncertainty and biases in experimental ML models
• Combine processing and characterization data for property prediction
• Critically evaluate claims about ML in materials science

Author

Philipp Pelz
Materials Science and Engineering
Course Instructor & Content Development

License

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

References

Course materials and references are maintained in references.bib using BibTeX format.