Appendix D
Course Map
Where each topic of three widely used Stanford courses (CS229 Machine Learning, CS231n Deep Learning for Computer Vision, and CS230 Deep Learning) is covered in this book.
This book is independent and not affiliated with Stanford University or its courses. The tables below map the topics listed on the public course pages (https://cs229.stanford.edu, https://cs231n.stanford.edu, https://cs230.stanford.edu) to the chapters, sections, and projects that cover them, so that a reader following one of these courses can find the matching material. Course syllabi change from year to year; the mapping follows the 2025 offerings.
D.1CS229: Machine Learning
| Course topic | Where in this book |
|---|---|
| Supervised learning setup | Section 1.3 |
| Linear regression, normal equations, LMS | Chapter 4; Project 1 |
| Probabilistic interpretation, locally weighted regression | Sections 4.4 and 4.7 |
| Logistic regression, perceptron, Newton’s method | Chapter 6, Sections 6.6, 6.7, and 3.8 |
| Exponential family and generalized linear models | Chapter 7 |
| Generative learning: GDA, Naive Bayes, Laplace smoothing | Sections 8.5 and 8.4; Project 2 |
| Kernels and support vector machines | Chapter 9; Project 8 |
| Neural networks and backpropagation | Chapter 15; Project 3 |
| Bias–variance, regularization, model selection, cross-validation | Chapters 5 and 12 |
| Learning theory: sample complexity, VC dimension | Chapter 11 |
| Double descent and the modern view of generalization | Sections 5.6 and 11.7 |
| Decision trees, bagging, boosting | Chapter 10 |
| $k$-means, mixtures of Gaussians, EM | Chapter 13; Project 4 |
| Factor analysis | Section 13.8 |
| PCA and ICA | Chapter 14, Section 14.8; Project 5 |
| Self-supervised learning and foundation models | Chapters 26 and 24 |
| MDPs, value iteration, policy iteration, model learning | Sections 27.1 and 27.2 |
| Continuous states, fitted value iteration | Section 27.3 |
| LQR, DDP, LQG | Section 27.4 |
| Policy gradient (REINFORCE) | Section 27.6 |
| Advice for applying machine learning | Chapters 12 and 20 |
D.2CS231n: Deep Learning for Computer Vision
| Course topic | Where in this book |
|---|---|
| Image classification, $k$-nearest neighbors | Section 17.1, Chapter 8; Project 9 |
| Linear classifiers, SVM and softmax losses | Sections 17.2 and 9.8; Project 9 |
| Regularization and optimization | Chapters 3 and 5, Section 16.4 |
| Neural networks and backpropagation | Chapter 15; Project 9 |
| Convolutional networks | Sections 17.3, 17.4, and 17.5 |
| CNN architectures and training | Sections 17.6 and 17.7, Chapter 16 |
| Recurrent networks, LSTMs, image captioning | Chapter 18; Project 10 |
| Attention and Transformers, Vision Transformers | Chapter 19, Section 19.9 |
| Object detection and segmentation | Chapter 21; Project 11 |
| Visualizing and understanding, adversarial examples, style transfer | Chapter 22; Project 12 |
| Video understanding | Section 23.1 |
| Large-scale distributed training | Section 20.6 |
| Self-supervised learning | Chapter 26 |
| Generative models: VAEs, GANs, autoregressive, diffusion, flows | Chapter 25; Project 7 |
| 3D vision: representations, NeRF, Gaussian splatting | Sections 23.2 and 23.3 |
| Vision and language | Sections 23.4 and 26.6 |
| Robot learning | Section 27.8, Chapter 27 |
D.3CS230: Deep Learning
| Course topic | Where in this book |
|---|---|
| Neural network basics, vectorization | Chapter 15, Appendix A |
| Initialization, regularization, batch normalization | Sections 16.1, 16.3, and 16.4 |
| Optimization algorithms, hyperparameter tuning | Chapter 3, Section 16.6 |
| Structuring projects: metrics, splits, error analysis | Sections 20.1, 20.2, and 20.3 |
| Mismatched data, transfer, multitask, end-to-end learning | Sections 20.4 and 20.5 |
| Convolutional and recurrent networks | Chapters 17 and 18 |
| Adversarial examples and GANs | Section 22.4, Section 25.4 |
| Interpretability | Chapter 22 |
| Deep reinforcement learning | Sections 27.5 and 27.6 |
| Deployment and efficient inference | Section 20.7 |
D.4Reading Paths
- A machine learning course (CS229-style). Parts I–III, then Chapter 15, Chapter 26, and Chapter 27, with Projects 1–5 and 8.
- A computer vision course (CS231n-style). Chapters 3, 4, and 6 as needed, then Part IV and Part V, Chapters 25 and 26, and Section 27.8, with Projects 3, 9, 10, 11, and 12.
- A practical deep learning course (CS230-style). Chapters 15, 16, 17, 18, 20, and 22, with Projects 3, 6, and 9.