Lectures 9-13


These lectures covered primarily the following four papers:

  1. Multi-way spectral partitioning and higher-order Cheeger inequalities
  2. Subexponential Algorithms for Unique Games and Related Problems
  3. Approximating the Expansion Profile and Almost Optimal Local Graph Clustering
  4. Finding Small Sparse Cuts Locally by Random Walk

I will post comprehensive lecture notes soon. We only used the subspace enumeration algorithm from (2) and the bulk of our analysis used the Lovasz-Siminovits-style arguments of (3) (see also (4)).