diff options
| -rw-r--r-- | finale/mid_report.tex | 14 |
1 files changed, 12 insertions, 2 deletions
diff --git a/finale/mid_report.tex b/finale/mid_report.tex index 684d0a8..e35a839 100644 --- a/finale/mid_report.tex +++ b/finale/mid_report.tex @@ -35,7 +35,7 @@ \newtheorem*{example}{Example} \newtheorem*{remark}{Remark} -\title{Regression Analysis with Network Data} +\title{Network Inference from Cascades} \author{Thibaut Horel \and Jean Pouget-Abadie} \begin{document} @@ -43,7 +43,17 @@ \maketitle \begin{abstract} - + The Network Inference Problem (NIP) is the machine learning challenge of + recovering the edges and edge weights of an unknown weighted graph from the + observations of a random contagion process propagating over this graph. + While estimators with provable convergence rate guarantees have been + obtained under various formulations of the NIP, a rigorous statistical + treatment of the problem is still lacking. In this work, we build upon the + unified NIP formulation of [] to explore the connections between the + topological properties of the graph to be learnt and the resulting quality + of the estimators. Specifically, we analyze which properties of the graph + render NIP unfeasible or hard, and which properties can be exploited to + improve the quality of the estimators. \end{abstract} \section{Introduction} |