Course positioning
CS 229 is one of Stanford's most classic and influential machine-learning courses, and it goes noticeably deeper than CS 230. It is less about how to call a model and more about the assumptions, losses, optimization, and generalization behind it — the foundation for research or a PhD.
What it covers
- Supervised learning: linear/logistic regression, GLMs, SVMs, kernels, trees and ensembles
- Learning theory: bias-variance, regularization, feature selection, generalization and model selection
- Unsupervised learning: k-means, GMMs, EM, PCA, ICA, density estimation
- Deep learning basics: neural networks, backpropagation, training techniques
- Reinforcement learning: MDPs, value iteration, policy search, adaptive control
- Course project: usually an open-ended ML research or application project
Prerequisites
Python/NumPy programming (CS 106A/B level); probability (CS 109/MATH 151/STATS 116); multivariable calculus and linear algebra (MATH 51/CS 205).
Assessment & workload
Typically 4–5 problem sets plus a midterm and a final project; about 10–15 hours per week. Check the current syllabus for exact weights.
Who it's for
People who want to understand the math behind ML and later do research or a PhD; engineers who already use scikit-learn but want the derivations and theory; interview candidates who want more than memorized answers.
Study advice
Write the problem-set derivations yourself, especially GLMs, the SVM dual, and EM. Pair it with CS 229M or STATS 315A to fill in the theory side.
Related courses
Official course description
Topics: statistical pattern recognition, linear and non-linear regression, non-parametric methods, exponential family, GLMs, support vector machines, kernel methods, deep learning, model/feature selection, learning theory, ML advice, clustering, density estimation, EM, dimensionality reduction, ICA, PCA, reinforcement learning and adaptive control, Markov decision processes, approximate dynamic programming, and policy search. Prerequisites: knowledge of basic computer science principles and skills at a level sufficient to write a reasonably non-trivial computer program in Python/NumPy to the equivalency of CS106A, CS106B, or CS106X, familiarity with probability theory to the equivalency of CS 109, MATH151, or STATS 116, and familiarity with multivariable calculus and linear algebra to the equivalency of MATH51 or CS205.