Lecture Notes on the ARV Algorithm for Sparsest Cut.

One of the landmarks in approximation algorithms is the $O(sqrt{log n})$-approximation algorithm for the Uniform Sparsest Cut problem by Arora, Rao and Vazirani from 2004. The algorithm is based on a semidefinite program that finds an embedding of the nodes respecting the triangle inequality. Their core argument shows that a random hyperplane approach will find two large sets of $Theta(n)$ many nodes each that have a distance of $Theta(1/sqrt{log n})$ to each other if measured in terms of $|cdot |_2^2$. Here we give a detailed set of lecture notes describing the algorithm. For the proof of the Structure Theorem we use a cleaner argument based on expected maxima over $k$-neighborhoods that significantly simplifies the analysis.