About this Lecture

TL;DR: This course gives an overview of both classic machine learning (linear regression/classification, GMM, SVM, GP) as well as more modern deep learning methods (MLP, CNN, RNN, AE). The only prerequisites are linear algebra and some basics of probability theory. After completing the course, you will know how to choose the appropriate class of supervised/unsupervised learning algorithms, how to train a model, and how to evaluate it.

Welcome to the 4th edition of the Introduction to Scientific Machine Learning for Engineers in the winter semester 2024/2025! We look forward to a hopefully great semester and to excite as many of you as possible for Scientific Machine Learning.

The course breaks down into an introduction to the topic, followed by the core content (“Lecture”), which is interspersed with practice problems while being supported by JuPyter notebook-based tutorials (“Exercise”) for the practical application of the learned concepts.

Lecturers

• Nikolaus A. Adams

• Artur Toshev (artur.toshev@tum.de)

• Harish Ramachandran (harish.ramachandran@tum.de)

Questions should preferably be posted on Moodle or sent to Artur Toshev or Harish Ramachandran.

Outline

• Introduction

Lecture

• 1. Linear Models
• 2. Gaussian Mixture Models
• 3. Bayesian methods
• 4. Optimization
• 5. Tricks of Optimization
• 6. Support Vector Machines
• 7. Gaussian Processes
• 8. Gradients
• 9. Multilayer Perceptron
• 10. Convolutional Neural Networks
• 11. Recurrent Models
• 12. Encoder-Decoder Models

Exercise

• 1. Linear and Logistic Regression
• 2. Bayesian Linear Models
• 3. Optimization
• 4. Support Vector Machines
• 5. Gaussian Processes
• 6. Convolutional Neural Networks
• 7. Recurrent Neural Networks

Miscellaneous

• Admin
• Books
• References
• Preliminary Knowledge
• Software Infrastructure
• Practical Exam WS22/23
• Frequently Asked Questions

Contributors

Thanks to all contributors! Github names are provided in parentheses where available.

Contributed content

1. Armin Illerhaus - Notebooks on Windows

Content fixes

1. Andreas Steger (AndSte01)

2. Muhammet Ali Güldali

Citation

Please cite this work as:

@article{paehler2024sciml,
  title={Introduction to Scientific Machine Learning for Engineers},
  author={Ludger Paehler and Artur P Toshev and Nikolaus A Adams},
  url={https://tumaer.github.io/SciML/about.html},
  year={2024},
}