• Corpus ID: 115173011

Concentration of the adjacency matrix and of the Laplacian in random graphs with independent edges

@article{Oliveira2010ConcentrationOT,
  title={Concentration of the adjacency matrix and of the Laplacian in random graphs with independent edges},
  author={Roberto Imbuzeiro Oliveira},
  journal={arXiv: Combinatorics},
  year={2010}
}
  • R. Oliveira
  • Published 3 November 2009
  • Mathematics, Computer Science
  • arXiv: Combinatorics
Consider any random graph model where potential edges appear independently, with possibly different probabilities, and assume that the minimum expected degree is !(lnn). We prove that the adjacency matrix and the Laplacian of that random graph are concentrated around the corresponding matrices of the weighted graph whose edge weights are the probabilities in the random model. We apply this result to two different settings. In bond percolation, we show that, whenever the minimum expected degree… 
On the Spectra of General Random Graphs
TLDR
A Chernoff inequality for matrices is used to show that the eigenvalues of the adjacency matrix and the normalized Laplacian of such a random graph can be approximated by those of the weighted expectation graph, with error bounds dependent upon the minimum and maximum expected degrees.
Concentration and regularization of random graphs
TLDR
This paper studies how close random graphs are typically to their expectations through the concentration of the adjacency and Laplacian matrices in the spectral norm, based on Grothendieck‐Pietsch factorization, using which a new decomposition of random graphs is constructed.
Concentration of the Stationary Distribution on General Random Directed Graphs
We consider a random model for directed graphs whereby an arc is placed from one vertex to another with a prescribed probability which may vary from arc to arc. Using perturbation bounds as well as
Title Concentration and regularization of random graphs Permalink
TLDR
This paper studies how close random graphs are typically to their expectations through the concentration of the adjacency and Laplacian matrices in the spectral norm, based on GrothendieckPietsch factorization, using which a new decomposition of random graphs is constructed.
Spectral Clustering of Graphs with General Degrees in the Extended Planted Partition Model
TLDR
A spectral clustering algorithm for similarity graphs drawn from a simple random graph model, where nodes are allowed to have varying degrees, is examined, and guarantees on the performance are shown that it outputs the correct partition under a wide range of parameter values.
Asymptotics of the spectral radius for directed Chung-Lu random graphs with community structure
TLDR
The spectral radius of the adjacency matrix can impact both algorithmic efficiency as well as the stability of solutions to an underlying dynamical process and this work provides novel concentration results for the spectral radius for the directed Chung-Lu random graph model.
Sparse random graphs: regularization and concentration of the Laplacian
TLDR
It is proved that this regularization indeed forces Laplacian to concentrate even in sparse graphs, establishing the validity of one of the simplest and fastest approaches to community detection -- regularized spectral clustering, under the stochastic block model.
On the Spectra of General Random Mixed Graphs
TLDR
A new probability inequality is derived and applied to obtain an upper bound on the eigenvalues of the Hermitian adjacency matrix, and the eigens of the normalized Hermitia Laplacian matrix can be approximated by a closely related weighted expectation matrix, with error bounds depending on the minimum expected degree of the underlying undirected graph.
Limit theorems for eigenvectors of the normalized Laplacian for random graphs
We prove a central limit theorem for the components of the eigenvectors corresponding to the $d$ largest eigenvalues of the normalized Laplacian matrix of a finite dimensional random dot product
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 66 REFERENCES
The emergence of a giant component in random subgraphs of pseudo‐random graphs
TLDR
It is shown that if $\alpha 1$ then G_p contains a unique giant component of size $\Omega(n)$, with all other components of size $O(\log n)$.
The Cut Metric, Random Graphs, and Branching Processes
TLDR
The component structure of random graphs with independence between the edges is studied, and whether there is a giant component is determined, and its asymptotic size is found when it exists.
The Spectral Gap of a Random Subgraph of a Graph
TLDR
It is shown that a random subgraph of G on n vertices with edge-selection probability p almost surely has as its spectral gap the gap between its normalized Laplacian and minimum degree d min.
Percolation on finite graphs and isoperimetric inequalities
Consider a uniform expanders family Gn with a uniform bound on the degrees. It is shown that for any p and c>0, a random subgraph of Gn obtained by retaining each edge, randomly and independently,
The Spectral Gap of Random Graphs with Given Expected Degrees
TLDR
The main result is that w.h.p. G(n,d) has a large subgraph core(G(n),d) such that the spectral gap of the normalized Laplacian of core(N,d)) is $\geq1-c_0{\bar d}_{\min}^{-1/2}$ with high probability; here c0>0 is a constant, and ${\bar d}.
Spectral graph theory
  • U. Feige
  • Mathematics
    Zeta and 𝐿-functions in Number Theory and Combinatorics
  • 2019
With every graph (or digraph) one can associate several different matrices. We have already seen the vertex-edge incidence matrix, the Laplacian and the adjacency matrix of a graph. Here we shall
Random Graph Coverings I: General Theory and Graph Connectivity
TLDR
A simple model for random graphs that have an n-fold covering map onto a fixed finite base graph and the main specific result is that if is the smallest vertex degree in G, then almost all n-covers of G are -connected.
Partitioning Random Graphs with General Degree Distributions
TLDR
A natural variant of this model that features a planted partition is studied, the main result being that there is a polynomial time algorithm for recovering (a large share of) the planted partition efficiently.
Random subgraphs of finite graphs: I. The scaling window under the triangle condition
TLDR
Among the models for which the triangle diagram is small enough to allow us to draw these conclusions are the random graph, the n‐cube and certain Hamming cubes, as well as the spread‐out n‐dimensional torus for n > 6.
Spectral techniques applied to sparse random graphs
  • U. Feige, E. Ofek
  • Mathematics, Computer Science
    Random Struct. Algorithms
  • 2005
TLDR
The eigenvalue gap for the adjacency matrices of sparse random graphs is analyzed and it is shown that by removing a small number of vertices of highest degree in G, one gets a graph G′ for which $\lambda = O(\sqrt{d})$.
...
1
2
3
4
5
...