Introduction to Scientific Machine Learning (Lecture Book)
Author: Ilias Bilionis (Purdue University, USA)
About this book
These online lecture notes accompany ME 539 (Introduction to Scientific Machine Learning). They are designed to complement the live lectures rather than serve as a standalone textbook. The course takes a probability-first approach to scientific machine learning: uncertainty as a modeling language, Monte Carlo methods for uncertainty propagation, and core supervised and unsupervised learning concepts. It also introduces state space models, physics-informed deep learning, and automated Bayesian inference via probabilistic programming.
The material is hands-on and Python-based, with practical examples that connect data-driven methods to engineering and scientific modeling.
Contents
- Introduction
- Review of Probability
- Uncertainty Propagation
- Principles of Bayesian Inference
- Supervised Learning
- Unsupervised Learning
- State Space Models
- Gaussian Process Regression
- Neural Networks
- Advanced Methods for Characterizing Posteriors
- Homework
Readership
For advanced undergraduates, graduate students, and practitioners working with uncertainty-aware models.
Explore more Books
Browse the full PSL book collection—published titles and online textbooks.