[Tex/LaTex] Style Pseudocode in LaTeX

Question

I'm trying to replicate the style of the following pseudocode taken from 'The Elements of Statistical Learning'.

I've tried with algorithm,algorithmic and enumerate (for the item) but I haven't reached the result below…

Answer 1

This is just a nested enumerate inside an algorithm environment:

\documentclass{report}

\usepackage{algorithm,amsmath,lipsum,xcolor,caption}
\DeclareMathOperator{\sgn}{sign}
\DeclareCaptionFont{lightblue}{\color{blue}}
\captionsetup[algorithm]{labelfont={bf,lightblue}, textfont={it}}
\renewcommand{\thealgorithm}{\thechapter.\arabic{algorithm}}

\begin{document}

\setcounter{chapter}{10}% Just for this example
\lipsum[1]

\begin{algorithm}[htb]
  \caption{AdaBoost.M1.}
  \begin{enumerate}
    \item
    Initialize the observation weights $w_i = 1/N$, $i = 1, 2, \dots, N$

    \item
    For $m = 1$ to $M$:
    \begin{enumerate}
      \item
      Fit a classifier $G_m(x)$ to the training data using weights $w_i$.

      \item
      Compute
      \[
        \text{err}_m = \frac{\sum_{i=1}^N w_i I(y_i \neq G_m(x_i))}{\sum_{i=1}^N w_i}.
      \]

      \item
      Compute $\alpha_m = \log((1 - \text{err}_m) / \text{err}_m)$.

      \item
      Set $w_i \leftarrow w_i \cdot \exp [\alpha_m \cdot I(y_i \neq G_m(x_i))]$, $i = 1, 2, \dots, N$.
    \end{enumerate}

  \item
  Output $G(x) = \sgn \bigl[ \sum_{m = 1}^M \alpha_m G_m(x) \bigr]$.
  \end{enumerate}
\end{algorithm}

\lipsum[2]

\end{document}