ridge regression as SVD
(1.1 hours to learn)
Summary
It's possible to write the ridge regression solution in terms of the SVD of the dataset. This gives insight into how it makes predictions. It also gives a way of defining the "degrees of freedom" or "effective number of parameters" of the model, which lets us analyze the degree of overfitting.
Context
This concept has the prerequisites:
Core resources (read/watch one of the following)
-Free-
→ The Elements of Statistical Learning
A graudate-level statistical learning textbook with a focus on frequentist methods.
-Paid-
→ Machine Learning: a Probabilistic Perspective
A very comprehensive graudate-level machine learning textbook.
Location:
Section 7.5.3, pages 228-230
Other notes:
- If you don't know what PCA is, just think of it as the SVD.
See also
-No Additional Notes-