Abstract
We give a new algorithm for Unique Games which is based on purely spectral techniques, in contrast to previous work in the area, which relies heavily on semidefinite programming (SDP). Given a highly satisfiable instance of Unique Games, our algorithm is able to recover a good assignment. The approximation guarantee depends only on the completeness of the game, and not on the alphabet size, while the running time depends on spectral properties of the Label-Extended graph associated with the instance of Unique Games. We further show that on input the integrality gap instance of Khot and Vishnoi, our algorithm runs in quasi-polynomial time and decides that the instance is highly unsatisfiable. Notably, when run on this instance, the standard SDP relaxation of Unique Games fails. As a special case, we also re-derive a polynomial time algorithm for Unique Games on expander constraint graphs. The main ingredient of our algorithm is a technique to effectively use the full spectrum of the underlying graph instead of just the second eigenvalue, which is of independent interest. The question of how to take advantage of the full spectrum of a graph in the design of algorithms has been often studied, but no significant progress was made prior to this work.
Original language | English (US) |
---|---|
Pages (from-to) | 177-206 |
Number of pages | 30 |
Journal | Computational Complexity |
Volume | 20 |
Issue number | 2 |
DOIs | |
State | Published - Jun 2011 |
Externally published | Yes |
Keywords
- Unique games
- algorithms
- graph spectra
ASJC Scopus subject areas
- Theoretical Computer Science
- General Mathematics
- Computational Theory and Mathematics
- Computational Mathematics