Some worst-case datasets of deterministic first-order methods for solving binary logistic regression

We present in this paper some worst-case datasets of deterministic first-order methods for solving large-scale binary logistic regression problems. Under the assumption that number algorithm iterations is much smaller than problem dimension, with our it requires at least \begin{document}$ {{{\mathcal O}}}(1/\sqrt{\varepsilon}) $\end{document} oracle inquiries to compute an id="M2">\begin{document}$ \varepsilon $\end{document}-approximate solution. From traditional iteration complexity analysis point view, loss functions are new function instances among class smooth convex optimization