Reorganize section 2

1 files changed, 8 insertions, 13 deletions

diff --git a/project2/main.tex b/project2/main.tex
index 8b29b5b..daf7411 100644
--- a/project2/main.tex
+++ b/project2/main.tex
@@ -123,10 +123,6 @@ for the multi-agent problem which is a $\gamma\cdot\alpha$ approximation to the
 revenue-optimal mechanism where $\gamma$ is a constant which is at least
 $\frac{1}{2}$.
 
-
-To provide some more intuition about this, we assume that the buyer is charged a price $p_0$ to participate in the mechanism, 
-and then is offered a menu of goods with prices $p_1,...,p_m$. The buyer's utility over the goods is additive, as above. However, there is an ex-ante constraint of being allocated a given good $i$, given by $\hat{x}_i$. For each good, if the buyer is allocated the good, which he is with probability $x_i \leq \hat{x}_i$, then he pays $p_i$; otherwise, he pays nothing. This is essentially the concept of a two-part tariff, as discussed in \cite{armstrong}.
-
 It is interesting to consider two specific cases for which we already have an
 answer to our problem:
 \begin{itemize}
@@ -141,15 +137,14 @@ answer to our problem:
         which TODO:cite provides a 2 approximation to $R\left( \left(\frac{1}{m},\ldots,\frac{1}{m}\right)\right)$.
 \end{itemize}
 
-\section{Related Work}
-
-In \cite{babaioff}, the authors describe a setting with a monopolist seller,
-offering $n$ heterogeneous goods, and a single buyer. 
-\cite{hart}
-\cite{hartline}
-\cite{yao} 
-\cite{armstrong}
-\cite{alaei}
+In the general case, a candidate simple mechanism suggested by Jason Hartline
+is the following: the buyer is charged a price $p_0$ to participate in the
+mechanism, and then is offered a menu of goods with prices $p_1,...,p_m$.
+However, there is an ex-ante constraint of being allocated a given good $i$,
+given by $\hat{x}_i$.  For each good, if the buyer is allocated the good, which
+he is with probability $x_i \leq \hat{x}_i$, then he pays $p_i$; otherwise, he
+pays nothing. This is essentially the concept of a two-part tariff, as
+discussed in \cite{armstrong}.
 
 \bibliographystyle{abbrv}
 \bibliography{main}