Course Contents#
Introduction
Motivations and scope
Brief history of AI in science
Expectations and assessments
Exercise: Scientific programming
Machine Learning Basics
Concepts and terminology
Learning by example
Supervised
Unsupervised
Reinforcement
Exercise: Crystal hardness
Materials Data
Data sources and types
FAIR data principles
Data quality control
Exercise: Data-driven thermoelectrics
Crystal Representations
Composition
Structure
Crystal Graphs
Exercise: Navigating crystal space
Classical Learning
k-nearest neighbours
k-means clustering
Decision trees and beyond
Exercise: Metal or insulator?
Deep Learning
From neuron to perceptron
Network architecture and training
Convolutional neural networks
Exercise: Learning microstructure
Building a Model from Scratch
Data preparation
Model choice
Training and testing
Exercise: Crystal hardness II
Accelerated Discovery
Robotics and self-driving laboratories
Bayesian optimisation
Reinforcement learning
Exercise: Closed-loop optimisation
Generative Artificial Intelligence
Large language models
From latent space to diffusion
Agentic research
Exercise: Research challenge
Future Directions
Guest lecture
Exercise: Research challenge