Further reading

Where the rigor continues when a page points beyond this site. Per-page references appear in each note's Going further; these are the recurring book-length sources.

• Shai Shalev-Shwartz & Shai Ben-David, Understanding Machine Learning: From Theory to Algorithms, Cambridge, 2014. The standard rigorous treatment of the learning framework, perceptron, SVM, and generalization theory. Freely available from the authors.
• Christopher Bishop, Pattern Recognition and Machine Learning, Springer, 2006. The probabilistic view: likelihood, logistic regression, kernels.
• Trevor Hastie, Robert Tibshirani & Jerome Friedman, The Elements of Statistical Learning, 2nd ed., Springer, 2009. The statistical view: risk, validation, model assessment. Freely available from the authors.
• Stephen Boyd & Lieven Vandenberghe, Convex Optimization, Cambridge, 2004. Everything this site states about convexity, duality, and descent methods, in full. Freely available from the authors.
• Nicholas Higham, Accuracy and Stability of Numerical Algorithms, 2nd ed., SIAM, 2002. Floating point and error analysis, definitively.
• Bernhard Schölkopf & Alexander Smola, Learning with Kernels, MIT Press, 2002. Kernels, RKHS theory, and the representer theorem at depth.