Derivation of Ridge and Lasso Regularization from Bayesian Principles

This lesson explores how Ridge and Lasso regularization methods can be derived using Bayesian principles, starting from Bayes' rule and leading to point estimation by maximizing the log posterior distribution.

crotoneacademia•549 views•Jan 19, 2021•14:02

🔥 Related Trending Topics

LIVE TRENDS

This video may be related to current global trending topics. Click any trend to explore more videos about what's hot right now!

THIS VIDEO IS TRENDING!

This video is currently trending in Germany under the topic 'ted lasso'.

Trending Now Globally

bridget maasland

สภาพอากาศ

farul constanța - botoşani

الطقس غدًا

About this video

In this lesson, we start from the Bayes rule and show how it reduces to a point estimation of parameters when maximized the log value of the posteior distribution. Finally we show how to get Ridge regularization through a Gaussian prior and Lasso regularization through a Laplace prior.

Video Information

Views

549

Total views since publication

Likes

17

User likes and reactions

Duration

14:02

Video length

Published

Jan 19, 2021

Release date

Quality

hd

Video definition

About the Channel

crotoneacademia

View channel →

Tags and Topics

This video is tagged with the following topics. Click any tag to explore more related content and discover similar videos:

#machine learning #data science #regularizatoin #regression #lasso #ridge #Maxium a priori #MAP #bayes rule #bayesian learning #probablity #mathematics

Tags help categorize content and make it easier to find related videos. Browse our collection to discover more content in these categories.