Let the L2 regularized logistic regression function is given by, \begin{align} f(w) &= \frac{1}{N} \sum_i \log \left(1 + \exp\left\{ -t_i \left(w^T x_i\right)\right\} \right) + \mu \|w \|_2^2 \ = \frac{1}{N} \sum_i f_i(w), \end{align} where $t_i \in \mathbb{R}$, $w, x_i \in \mathbb{R}^n$, $\mu \in \mathbb{R}$, and $f_i(w) := \log \left(1 + \exp\left\{ -t_i \left(w^T x_i\right)\right\} \right) + \mu \|w \|_2^2$ .


  • How would/could I find the (small) Lipschitz constant $M$ and $\nu$-strongly parameter of $f(w)$?
  • How can I find the (small) Lipschitz constant $L$ of $\nabla f(w)$ and $L_i$ of $\nabla f_i(w)$?

I am sorry if this question has already been asked or it is trivial to compute (analytically?).


Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.