diff options
Diffstat (limited to 'paper/paper.tex')
| -rw-r--r-- | paper/paper.tex | 141 |
1 files changed, 141 insertions, 0 deletions
diff --git a/paper/paper.tex b/paper/paper.tex new file mode 100644 index 0000000..0a11443 --- /dev/null +++ b/paper/paper.tex @@ -0,0 +1,141 @@ +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +%%%%%%%% ICML 2015 EXAMPLE LATEX SUBMISSION FILE %%%%%%%%%%%%%%%%% +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% + +% Use the following line _only_ if you're still using LaTeX 2.09. +%\documentstyle[icml2015,epsf,natbib]{article} +% If you rely on Latex2e packages, like most moden people use this: +\documentclass{article} + +% use Times +\usepackage{times} +% For figures +\usepackage{graphicx} % more modern +%\usepackage{epsfig} % less modern +\usepackage{subfigure} + +% For citations +\usepackage{natbib} + +% For algorithms +\usepackage{algorithm} +\usepackage{algorithmic} + +% As of 2011, we use the hyperref package to produce hyperlinks in the +% resulting PDF. If this breaks your system, please commend out the +% following usepackage line and replace \usepackage{icml2015} with +% \usepackage[nohyperref]{icml2015} above. +\usepackage{hyperref} + +% Packages hyperref and algorithmic misbehave sometimes. We can fix +% this with the following command. +\newcommand{\theHalgorithm}{\arabic{algorithm}} + +% Employ the following version of the ``usepackage'' statement for +% submitting the draft version of the paper for review. This will set +% the note in the first column to ``Under review. Do not distribute.'' +\usepackage{icml2015} + +% Employ this version of the ``usepackage'' statement after the paper has +% been accepted, when creating the final version. This will set the +% note in the first column to ``Proceedings of the...'' +%\usepackage[accepted]{icml2015} + + +% The \icmltitle you define below is probably too long as a header. +% Therefore, a short form for the running title is supplied here: +\icmltitlerunning{Cracking Cascades: Sparse Recovery for Graph Inference} + +\begin{document} + +\twocolumn[ +\icmltitle{Cracking Cascades: Sparse Recovery for Graph Inference} + +% It is OKAY to include author information, even for blind +% submissions: the style file will automatically remove it for you +% unless you've provided the [accepted] option to the icml2015 +% package. +\icmlauthor{Your Name}{email@yourdomain.edu} +\icmladdress{Your Fantastic Institute, + 314159 Pi St., Palo Alto, CA 94306 USA} +\icmlauthor{Your CoAuthor's Name}{email@coauthordomain.edu} +\icmladdress{Their Fantastic Institute, + 27182 Exp St., Toronto, ON M6H 2T1 CANADA} + +% You may provide any keywords that you +% find helpful for describing your paper; these are used to populate +% the "keywords" metadata in the PDF but will not be shown in the document +\icmlkeywords{boring formatting information, machine learning, ICML} + +\vskip 0.3in +] + +\begin{abstract} +blabla +\end{abstract} + +\section{Introduction} + +\begin{itemize} +\item Small introduction about problem +\item What is a cascade? +\item What is our objective? +\item Motivation for the problem. +\item Summary of our approach +\end{itemize} + +Parameters of the model: +\begin{itemize} +\item $p_{init}$ : multi-source: explain why more reasonable +\item $p_{min}$ +\item $p_i <$ 1- epsilon? +\end{itemize} + +\section{Related Work} + +\subsection{Past work} + +\begin{itemize} +\item Gomez +\item Netrapalli: $d^2 $log n + correlation decay + very bad dependence on +\item Kleinberg/Abrahao: $d^9$ log n: single source model... +\item Gomez: $d^3$ log n + assumptions on Hessian of diffusion process: upper and lower bound on eigenvalues + same proof concept as Netrapalli +\end{itemize} + +\subsection{Our contribution} + +\begin{itemize} +\item Better assumptions: easy to understand, verify?, and much less restrictive +\item Oracle inequality rather than support recovery -> First one +\item Algorithm for recovery in Omega(d log n) -> First one +\item Practical Confidence Intervals +\item Practical Lower bound +\item Compare on generic networks +\end{itemize} + +To justify: +\begin{itemize} +\item why discrete isn't so bad; +\item why Gomez's assumptions are not reasonable; +\end{itemize} + +\section{Algorithm and Theoretical Guarantees} + +\subsection{Independent Cascade Model} + +\subsection{The Voter Model} + +\section{Matching the Lower Bound} + +\section{Practical Confidence Intervals} + +\section{Experiments} + +\section{Discussion} + + +\bibliography{example_paper} +\bibliographystyle{icml2015} + +\end{document} + |