Gradient Descent Equations - Mathematical Derivation [Machine Learning Tutorial]

preview_player
Показать описание
Although we don't require to prove this equation to use it in coding(just one line of code in Python) ,but proving makes lot of clarity in our mind if we wish to use any optimization method such as Gradient Descent.
Рекомендации по теме
Комментарии
Автор

That was an excellent explanation. I just wanted to add one tiny caveat... this all assumes that (1) your linear approximation is good enough I.e. it’s inherently an approximation, and (2) alpha is only an approximation off infinitesimal/abs(grad(f)). Realizing this makes things so much easier when you start doing ML (at least it did for me). Again, wonderful video!

ozzyfromspace
Автор

TQ SIR...WE EXPECT MORE VIDEOS FROM YOU SIR

hemanthmahi