Introduction to conditional probability, matrix representation of discrete probabilities, Bayes rule and examples, such as pagerank for web search and medical diagnosis.

The slides are available here:

This machine learning course was taught in 2012 at UBC by Nando de Freitas

## 12 comments

Comments are closed.

Thank You.. really helpful..

great lecture.

very nice lecture. really helpful

Thank you very much Professor De Freitas – It is extremely kind of you to

give us access to your video. Unfortunately, I am little bit old to tacke a

Phd now.

Thanks for the video. Is there any resource that describes the connection

between the second eigen value and the rate of convergence without all the

jargons related to special type of vectors and matrices.

I have been watching all your lectures and I am really amazed with the high

quality of them. Thanks for publishing it.

What is the G matrix here ?

Should you be looking to have higher rank on internet search engine, i

recommend using “googleranktool weebly com”. Just google it on search

engine.

where can we get the assignment solutions ?

“i’ve built search engines, i’ve sold search engines for shit loads of

money” best line from a professor during a lecture!!!

Can anyone explain how π3 = π2*G?

Why is G not having any suffix like G1, G2, G3, etc? What I see is that G1

is P(X2|X1), and G2 should be P(X3|X2)… Thus G seems to be changing

according to me. How can we then write PI1*G^(K-1) = PIK?