1
$\begingroup$

Logistic and Linear Regression have different cost functions. But I don't get how the gradient descent in logistic regression is the same as Linear Regression.

We get the Gradient Descent formula by deriving the Squared Error cost function. However in Logistic Regression we use a Logarithmic Cost function instead. I think I am lost here.

$\endgroup$
  • 1
    $\begingroup$ Have you tried to take derivative yourself? I mean are you familiar with derivative and chain rule? $\endgroup$ – Media Jan 22 '18 at 10:42
  • $\begingroup$ @Media I don't know advanced calculus but still it was weird to me that both the cost functions' derivative leads to the same formula $\endgroup$ – Huzo Jan 22 '18 at 12:13
  • 1
    $\begingroup$ take a look at here which may help you find all the stuff needed for ML. $\endgroup$ – Media Jan 22 '18 at 12:17
2
$\begingroup$

Gradient Descent is an universal method, you can us it with basically every loss function you can find in known ML algorithms.

In your case, you have only to derive the logarithmic cost function. You can find a detailed calculation at

https://math.stackexchange.com/questions/477207/derivative-of-cost-function-for-logistic-regression

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.