Skip to main content

Ctrl+K

Learning from data for physicists

Overview

1. Invitation to inductive inference
2. Introduction

Part I: Bayesian methods for scientific modeling

3. Overview of Part I
4. Inference and PDFs
5. Bayesian posteriors
6. Updating via Bayes' rule
7. Error propagation
8. Bayes in practice
9. Exercises for Part I

Part II: Advanced Bayesian methods

10. Overview of Part II
11. Discrepancy Models
12. Assigning probabilities
13. 📥 Dealing with outliers
- 13.10. Student t distribution from Gaussians
14. Bayes goes linear: History matching
15. Multi-model inference with Bayes
- 15.1. Model Selection
- 15.2. Model averaging and mixing

Part III: MCMC sampling

16. Overview of Part III
17. Intuition for MCMC
18. Details of MCMC
19. MCMC in practice
20. Advanced sampling

Part IV: Machine learning: A Bayesian perspective

21. Overview of Part IV
22. Overview of Gaussian processes
23. Logistic Regression
- 23.5. Machine Learning: First Examples
- 23.6. Exercise: Logistic Regression and neural networks
24. Bayesian Optimization
25. Machine learning: Overview
26. ANNs in the large-width limit (ANNFT)
- 26.11. 📥 Distributions of Randomly-Initialized ANNs
27. Bayesian neural nets
- 27.4. Demonstration: Variational Inference and Bayesian Neural Networks
- 27.5. Exercise: Bayesian neural networks
28. *Convolutional neural nets

Part V: Other topics

29. Overview of Part V
30. Emulators
31. PCA, SVD, and all that
- 31.5. demo-SVD notebook
32. QBism: Bayesian quantum mechanics

Backmatter

33. Bibliography
34. Guide to Jupyter Book markdown

Appendix A: Reference

35. Statistics concepts and notation
36. Gradient-descent optimization

Appendix B: Scientific modeling

37. Overview of scientific modeling material
38. Overview of modeling
39. Linear models
40. Mathematical optimization

Appendix C: Getting started

41. Overview of Getting started material
42. Exercise: Jupyter notebooks and Python
43. More about Python and Jupyter notebooks
44. Setting up for using this Jupyter Book
- 44.1. Using Anaconda
- 44.2. Using GitHub

TALENT mini-projects

Overview of mini-projects
MP I: Parameter estimation for a toy model of an EFT
MP IIa: Model selection basics
MP IIb: How many lines?
- Overview of Mini-project IIb: How many lines?
MP IIIa: Bayesian optimization
MP IIIb: Bayesian Neural Networks

Binder

Repository
Open issue

.ipynb
.md

Overview of modeling

38. Overview of modeling#

“All models are wrong, but some are useful.”

—George Box, in Robustness in Statistics (1979)

previous

37. Overview of scientific modeling material

next

38.1. Notation

By Christian Forssén, Dick Furnstahl, and Daniel Phillips

© Copyright 2025.