11d Machine Learning: Bayesian Linear Regression
Lecture on Bayesian linear regression. By adopting the Bayesian approach (instead of the frequentist approach of ordinary least squares linear regression) we can account for prior information and directly model the distributions of the model parameters by updating with training data.
Follow along with the demonstration workflow:
github.com/GeostatsGuy/Python...
Пікірлер: 24
THIS IS THE ONLY GOOD EXPLANATION OF THIS!!! thank you
Your explanation was explicit , thank you.
Great explanation. Thank you
Great explanation and channel!
Thank you very much.
Thank you so much!!!!
It was a great explanation, thank you very much! I was wondering if you could tell the world a little bit about Bayesian Model Selection. One more time, thanks a lot.
very clear ....Thank you
Cool video
How is the distributions of uncertainty in Bayesian linear regression, differ from the confidence intervals of parameters in a frequentist linear regression ?
YOU ARE GREAT
@GeostatsGuyLectures
3 жыл бұрын
Thank you Jakob. I hope the content is helpful!
Also, in 2:31, shouldn't the equation at the bottom have (b0 + b1*x) rather than having a minus sign ?
@GeostatsGuyLectures
3 жыл бұрын
Howdy shan19key. Good catch. I'll fix the lecture and this will be updated in the next iteration. Appreciation!
hello thank you for this video. I just have a question regarding the equation at 10:46 for the first term of the top part of the fraction shouldn't it be P(y|X,beta) instead of P(y,X|beta)?
14:05 - isn't it intractable because the model parameters beta (not the features X) are continuous?
Hi there thanks for your lectures I've benefited heaps from them. I wanted to ask what book you refer to in this video "hasty book on statistical learning" ? Kind regards
@xxTrashBasherxx
2 жыл бұрын
Hastie, Tibshirani, Friedman - Elements of Statistical Learning
Nice video ... Can you use Bayesian regression to model nonlinear data? Greetings from Colombia. Thanks
@ghifariadamfaza3964
Жыл бұрын
Yes, you can! After all, the "linear" term in linear regression refers to the linearity in the parameter, not the data.
13:07 Posterior term is wrong in the text. What is written in the text is likelihood. But otherwise thanks.
@GeostatsGuyLectures
3 жыл бұрын
Great catch, Amin. I'll add errata to comments, correct this and post in the update. I appreciate the assistance!
good content, too bad you've been using your considerable intellect to benefit big oil. Yuck. What a waste.