CS 229 Notes by Andrew Ng

This is a link to pdf’s of notes by Andrew Ng following CS 229. I published this because, at most of the websites, the notes are either not indexed properly or there is not enough description for each note, or it’s on github and doesn’t directly open a pdf in a new tab.

1. Supervised Learning
   • Linear Regression
     • LMS Algorithm
     • Normal Equation
     • Probablistic Interpretation
     • Locally weighted linear regression
   • Classification and Logistic Regression
     • Logistic Regression
     • Digression
     • Another algorithm for maximising
   • Generalized Linear Models
     • Exponential Family
     • Constructing GLMs
   • Generative Learning Algorithms
     • Gaussian Discriminate Analysis
     • Naive Bayes
   • Support Vector Machines
     • Margins
     • Notation
     • Functional and Geometric margins
     • Optional margin classifier
     • Lagrange Duality
     • Optimal margin classifiers
     • Kernels
     • Regularisation
     • SMO Algo
   • Learning Theory
     • Bias/Variance Tradeoff
     • Preliminaries
     • Case of finite H
     • Case of infinite H
   • Regularization and Model selection
     • Cross validation
     • Feature selection
     • Bayesian Stats and regularization
       • Perceptron and large margine classifiers
2. Unsupervised learning
   • K-means clustering
   • Mixtures of Gaussian and EM algo
   • EM Algo
     • Jensen’s Inequality
     • EM Algo
     • Mixture of Gaussians revisited
   • Factor Analysis
     • Restrictions on Sigma
     • Marginals and Conditionals of Gaussian
     • Factor Analysis model
     • EM for factor analysis
   • Principal Component Analysis
   • Independent Component Analysis
     • ICA ambiguities
     • Density and Linear Transformations
     • ICA Algorithm
   • Reinforcement Learning and Control
     • Markov Processes
     • Value iteration and Policy iteration
     • Learning a model for MDP
     • Continous state MDPs