Learning objectives

In the three weeks we have for machine learning in this course we aim to

• give you a basic overview over the different "flavors" of ML (supervised, unsupervised, reinforcement)
• give you an idea how ML can be used in chemistry
• give you a basic understanding over the workflow of supervised ML
• familiarize you with the basic tooling to (pandas, sklearn) so that you can build models yourself
• provide you with the tools (e.g., reading documentation) and resources (e.g., Kaggle) to continue your learning journey

At the end of the course, you should be equipped to understand the general ideas of papers that use ML in chemistry.

Requirements

In some parts of the lecture and the theory exercises we’ll assume some familiarity with basics of linear algebra. You can find many refreshers in the web, for instance these slides.

The main concepts you should be familiar with:

• Rank
• Transpose of matrix
• Inverse of matrix
• Linear dependence
• Linear equations

Organization

ML course organization

<aside> ⚠️ Note that the slides used for the videos will slightly differ from the ones we will use in class (you'll see a different course number and different examples). In the class we will also discuss briefly gradient descent and the geometric interpretation of the normal equations. This is not covered in the videos—but also not crucial for the hands-on/project.

</aside>