diff options
| -rw-r--r-- | paper/sections/results.tex | 4 |
1 files changed, 1 insertions, 3 deletions
diff --git a/paper/sections/results.tex b/paper/sections/results.tex index 0c2c7e5..f62e6f2 100644 --- a/paper/sections/results.tex +++ b/paper/sections/results.tex @@ -1,7 +1,5 @@ In this section, we exploit standard techniques in sparse recovery and leverage the simple nature of generalized linear cascades to address the standard problem of edge detection. We extend prior work by showing that edge weights of the graph can be recovered under similar assumptions. We extend these results by considering the non-exactly sparse case. -\subsection{Objectives} - As mentioned previously, our objective is twofold: \begin{enumerate} @@ -99,4 +97,4 @@ Using the inequality $\forall x>0, \; \log x \geq 1 - \frac{1}{x}$, we have $|\l In other words, finding an upper-bound for the estimation error of the `effective' parameters $\theta_{i,j} \defeq \log(1-p_{i,j})$ provides immediately an upper-bound for the estimation error of the true parameters $(p_{i,j})_{i,j}$. -\subsubsection{The Voter Model}
\ No newline at end of file +\subsubsection{The Voter Model} |