0
$\begingroup$

Using the Representer theorem we can define a approximation function with or without the bias term $b$. If $b$ is considered, how should it be computed when training the primal SVM with gradient descent? Does $b$ affects the generalization performances?

$\endgroup$

Your Answer

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

Browse other questions tagged or ask your own question.