Machine Learning and Quantum Alchemy
Short course providing a high-level introduction into machine learning and quantum alchemy methods. Target audience both BSc and MSc level. Requires basic programming skills (in any language). Developed at University of Kassel and Universidade Federal do ABC with support from CAPES.
Every day, we will start with a 30 min theory lecture, followed by exercise tasks and further material for self-study. Bring your own device, we only require an account for Google Colab, no other installation needed. If you prefer your own python-installation, please make these packages available: scikit-learn, pyscf, and jax.
All exercise tasks will revolve around (molecular) chemistry, but do not require particular theoretical chemistry knowledge. At the end of each day, we will solve and discuss the tasks together. All material will be provided on this page even after the course has finished.
Python refresher
Those who feel they might want a brief refresher on Python, you may prepare with the relevant python course prior to this course.
-
Machine Learning Basics
Covers fundamental principles and key metrics to reliably train models using
scikit-learn. -
Applied Machine Learning
Introduces some learning algorithms and discusses their function using
scikit-learn. -
Quantum Alchemy
Discusses perturbative treatment of quantum chemistry and how to run
pyscf. -
Differentiable Quantum Chemistry
Explores automatic differentiation using
jaxand applications to quantum chemistry.