Hey guys, we'll today continue with the machine learning videos and cover a technique of learning the likelihood- Maximum Likelihood Estimation. Enjoy!
I have a good understanding of the end goal of choosing a distribution for each independent observation (3:10), however I am a bit confused on why this works. If we are really choosing a distribution for one observation, wouldn't this just be a delta function since it has a 100% chance of theta fitting the datapoint exactly at that point and a 0% chance elsewhere? It seems like we are almost choosing the standard deviation for the function. Apologies, I would just like to fully understand the process.
Well explained with the proper formal notations that is rare to be seen, good job
Thank you very much!! ☺️
Very Informative and well presented. I do enjoy such theoretical content. Hope you continue making such videos.
I‘m really glad you enjoyed it! I‘ll do my best to produce content that you guys enjoy ☺️ This will hopefully also include more theoretical content.
Many thanks for this succinct explanation.
very good explaination, keep posting more content like this
Again a great video Thanks
Thanks ☺️
Damn you teach so well and surprisingly look so good too
haha thank you very much ☺️
You are amazing!
Doing my best 😊
I have a good understanding of the end goal of choosing a distribution for each independent observation (3:10), however I am a bit confused on why this works. If we are really choosing a distribution for one observation, wouldn't this just be a delta function since it has a 100% chance of theta fitting the datapoint exactly at that point and a 0% chance elsewhere? It seems like we are almost choosing the standard deviation for the function. Apologies, I would just like to fully understand the process.
Choosing the most likelihood estimation over estimators would be a good choice by forming a Best fit Chi squared model.
Not quite sure what you mean 🥸
@@borismeinardus the list of estimators and the covariance of best fit by normal distribution....in a factory.
yes please the video on bayesian parameter estimation also
Is on the list!
Can you make the parameter estimation video please
@2:07 It should be Parametric form?
I'd recommend adding captions. It's hard to understand what's being said.
Do you study at TU Berlin ? 😂
I do :)