This repository contains the notebooks for the Spring tutorial on the automated experiment in materials synthesis and microscopy. The tutorial covers the principles of Gaussian Processes and Bayesian Optimization, structured GP, invariant VAEs including conditional, joint, and semisupervised versions, deep kernel learning, and forensics/human in the loop interventions. Topics include
- Introduction to Gaussian Processes
- Bayesian Optimization based on GP
- Bayesian Inference
- Structured GP
- Bayesian Hypothesis Learning
- Gaussian Processes beyond 1D
- Linear dimensionality reduction methods
- (Invariant) Variational Autoencoders
- Semi-supervised, joint, and conditional VAE
- VAE for imaging and spectroscopy problems - I
- VAE for imaging and spectroscopy problems - II
- Introduction to Deep Kernel Learning
- DKL for scientific discovery: process optimization
- Interpretable and human in the loop DKL AE
Note that for several topics there are only presentations, and for others there are only Colabs (with the explanations and suggested excercises)