Title :
On the second eigenvalue and linear expansion of regular graphs
Author_Institution :
MIT Lab. for Comput. Sci., Cambridge, MA, USA
Abstract :
The authors investigate the relation between the second eigen-value and the linear expansion of regular graphs. The spectral method is the best currently known technique to prove lower bounds on the expansion. He improves this technique by showing that the expansion coefficient of linear-sized subsets of a k-regular graph G is at least k/2(1-√max(0,1-λ1(G)24k-4))- , where λ1(G) is the second largest eigenvalue of the graph. In particular, the linear expansion of Ramanujan graphs, which have the property that the second largest eigenvalue is at most 2√k-1, is at least (k/2)-. This improves upon the best previously known lower bound of 3(k-2)/8. For any integer k such that k-1 is prime, he explicitly constructs an infinite family of k-regular graphs Gn on n vertices whose linear expansion is k/2 and such that λ1(Gn)⩽√k-1+0(1). Since the graphs Gn have asymptotically optimal second eigenvalue, this essentially shows the (k/2) is the best bound one can obtain using the second eigenvalue method
Keywords :
computational geometry; eigenvalues and eigenfunctions; graph theory; Ramanujan graphs; asymptotically optimal second eigenvalue; expansion coefficient; linear expansion; regular graphs; second eigenvalue; spectral method; Circuits; Complexity theory; Computer science; Concurrent computing; Contracts; Eigenvalues and eigenfunctions; Graph theory; Laboratories; Polynomials; Sorting;
Conference_Titel :
Foundations of Computer Science, 1992. Proceedings., 33rd Annual Symposium on
Conference_Location :
Pittsburgh, PA
Print_ISBN :
0-8186-2900-2
DOI :
10.1109/SFCS.1992.267762