1
$\begingroup$

This post: When is logistic regression solved in closed form? describes that we must use nonlinear optimization methods to find the parameter estimates for logistic regression models. Does gradient descent count as one of these methods? I have not heard of gradient descent described as a nonlinear optimization method. And what other "nonlinear optimization" methods are there for calculating the weights in this case?

$\endgroup$
1

1 Answer 1

0
$\begingroup$

A logistic regression is a neural network with no hidden layers. Consequently, any method used for calculating the weights in a neural network is fair game for a logistic regression.

$\endgroup$
1
  • $\begingroup$ The only method I know to calculate weights in a neural network is gradient descent, so is that our only option? $\endgroup$
    – xojfqa
    Apr 5, 2022 at 23:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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