'Partial Derivative term in the Gradient Descent Algorithm
I'm learning the "Machine Learning - Andrew Ng" course from Coursera. In the lesson called "Gradient Descent", I've found the formula a bit complicated. The theorem is consist of "partial derivative" term.
The problem for me to understand the calculation of partial derivative term. Thus, later the term is calculated as
1/m * ∑ (h θ (x) − y(i) )²
My question is, "How did the 1/2m from the 'Cost Function' becomes 1/m while calculating the partial derivative inside the Gradient Descent theorem?"
Solution 1:[1]
Differentiation of x²
is 2x
.
Similarly, differentiation of ??(h ??(x) ? y(i) )²
is 2 * ??(h ??(x) ? y(i) )
.
Therefore, differentiation of 1/2m * ??(h ??(x) ? y(i) )²
is 1/m * ??(h ??(x) ? y(i) )
.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
Solution | Source |
---|---|
Solution 1 | AHMED AGHADI |