Skip to content

Commit

Permalink
Slides (cont.)
Browse files Browse the repository at this point in the history
  • Loading branch information
glouppe committed Oct 5, 2014
1 parent abff3e1 commit f787a4a
Show file tree
Hide file tree
Showing 2 changed files with 17 additions and 9 deletions.
Binary file added slides/figures/blackbox-open.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
26 changes: 17 additions & 9 deletions slides/slides.tex
Original file line number Diff line number Diff line change
Expand Up @@ -736,16 +736,14 @@ \section{Inplementing and accelerating random forests}

\begin{table}
\centering
\begin{tabular}{| c | c c c |}
\begin{tabular}{| c | c |}
\hline
& \textit{Best case} & \textit{Worst case} & \textit{Average case} \\
& \textit{Average case} \\
\hline
\hline
CART & $\Theta(pN\log^2 N)$ & $O(pN^2\log N)$ & $\Theta(pN\log^2 N)$ \\
Bagging & $\Theta(Mp\widetilde{N}\log^2 \widetilde{N})$ & $O(Mp\widetilde{N}^2\log \widetilde{N})$ & $\Theta(Mp\widetilde{N}\log^2 \widetilde{N})$ \\
RF & $\Theta(MK\widetilde{N}\log^2 \widetilde{N})$ & $O(MK\widetilde{N}^2\log \widetilde{N})$ & $\Theta(MK\widetilde{N}\log^2 \widetilde{N})$ \\
ETs & $\Theta(MKN\log N)$ & $\Theta(MKN^2)$ & $\Theta(MKN\log N)$ \\
PERT & $\Theta(MN\log N)$ & $\Theta(MN^2)$ & $\Theta(MN\log N)$ \\
CART & $\Theta(pN\log^2 N)$ \\
Random Forest & $\Theta(MK\widetilde{N}\log^2 \widetilde{N})$ \\
Extra-Trees & $\Theta(MKN\log N)$ \\
\hline
\end{tabular}
\caption{Time complexity for building forests of $M$ randomized trees. $N$ denotes the number of samples in ${\cal L}$, $p$ the number of input variables and $K$ the number of variables randomly drawn at each node. $\widetilde{N} = 0.632 N$.}
Expand Down Expand Up @@ -916,23 +914,33 @@ \section{Inplementing and accelerating random forests}

\section{Conclusions}

\begin{frame}{Conclusions}
\begin{frame}{Opening the black box}
\begin{itemize}

\item Random forests constitute one of the most {\color{blue} robust and
effective} machine learning algorithms for many problems.

\medskip

\item While simple in design and easy to use, random forests remain however
\begin{itemize}
{\color{red}
\item hard to analyze statistically,
\item non-trivial to interpret,
\item difficult to implement properly.
\item difficult to implement properly.}
\end{itemize}

\medskip

\item Through an in-depth re-assessment of the method, this dissertation has
proposed {\color{blue} original contributions} on these issues.

\end{itemize}

\begin{center}
\includegraphics[scale=0.5]{figures/blackbox-open.jpg}
\end{center}

\end{frame}

\begin{frame}{Future works}
Expand Down

0 comments on commit f787a4a

Please sign in to comment.