#1312 overall
9.520 Statistical Learning Theory and Applications
MIT OpenCourseWare
1 recommender · 1 thread · recommended in 1 year, 2012 · score 0.7
A graduate-level MIT course covering the theory and practical applications of statistical learning, including regularization, kernel methods, and generalization. The linked slides cover spectral methods and show how projection acts as regularization.
One commenter calls the whole course great and points to its slides for an intuitive explanation of why projection onto principal components aids generalization.
Machine Learning & AIadvancedfreevideo lectures#statistical learning#regularization#kernel methods#generalization#spectral methods#rosasco
Recommendations by year
20062026
Indeed, projection onto the principal subspace is a kind of regularization, which makes for better generalization.
This kind of makes sense intuitively. These slides go into more detail (the whole course is great), where it says "Projection regularizes!"
http://www.mit.edu/~9.520/spring09/Classes/class07_spectral....