Braess's paradox for the spectral gap in random graphs and delocalization of eigenvectors.
Ronen EldanMiklós Z. RáczTselil SchrammPublished in: Random Struct. Algorithms (2017)
Keyphrases
- random graphs
- laplacian matrix
- graph theoretic
- learning curves
- covariance matrix
- phase transition
- spectral clustering
- small world
- spectral analysis
- principal component analysis
- spectral methods
- attributed graphs
- complex networks
- undirected graph
- graph laplacian
- directed graph
- search algorithm
- normalized cut
- input image