Data-Driven Modeling: Regression

This lecture provides an introduction to data-driven modeling, focusing on regression. It explains the concept of establishing a relationship between input data and target properties by recognizing patterns in the data. The lecture covers linear regression as a basic data-driven model, discussing the adjustment of parameters to minimize loss. It also delves into the risks of purely inductive reasoning and the inclusion of physics-inspired concepts in machine learning. Furthermore, it explores the use of principal component analysis (PCA) for dimensionality reduction and ridge regression for establishing relationships between fingerprints and physical properties.