this is for holding javascript data
Antonio Coppola edited Introduction.tex
over 9 years ago
Commit id: 2ffaccd9d5b15e2240dd1049932c19ff6b1648d2
deletions | additions
diff --git a/Introduction.tex b/Introduction.tex
index 5d2bf08..db14173 100644
--- a/Introduction.tex
+++ b/Introduction.tex
...
\subsection{The OWL-QN Algorithm}
The L-BFGS method cannot be applied to problems
involving $L_1$ norms with an objective function of the
parameter vector, form $r(x) = C \cdot \lvert \lvert x \lvert \lvert _1 C = \cdor \sum_i \lvert x_i \lvert$, such as LASSO regression or $L_1$-penalized log-linear models, given the non-differentiability of the objective function at any point where at least one of the parameters is zero. The OWL-QN algorithm exploits the fact that $L_1$-regularized objective functions will still
Trying out the math:
\[
r(x) = C \cdot \lvert \lvert x\lvert \lvert _1 = C \cdot \sum_i \lvert x_i \lvert
\]