Ridge Regularization: An Essential Concept in Data Science

• Activity Number: 358 - 50 Years of Ridge Regression
• Type: Invited
• Date/Time: Thursday, August 12, 2021 : 12:00 PM to 1:50 PM
• Sponsor: Technometrics
• Author(s): Trevor JOHN Hastie*
• Companies: STANFORD UNIVERSITY
• Keywords: ridge; quadratic; double descent; kernel trick

Abstract

This talk is part of the celebration of the 50th anniversary of Hoerl and Kennard's 1970 Technometrics paper on Ridge Regression. Ridge has come up in many aspects of my own research: generalized additive models, elastic net, support-vector machines and most recently understanding the "double-descent" phenomenon in machine learning.This talk will touch on all of these, and focus on the last.