Estimating distance between an eigenvalue of a signed graph and the spectrum of an induced subgraph (Q6094710)
From MaRDI portal
scientific article; zbMATH DE number 7737606
Language | Label | Description | Also known as |
---|---|---|---|
English | Estimating distance between an eigenvalue of a signed graph and the spectrum of an induced subgraph |
scientific article; zbMATH DE number 7737606 |
Statements
Estimating distance between an eigenvalue of a signed graph and the spectrum of an induced subgraph (English)
0 references
14 September 2023
0 references
In this paper, the authors are interested in determining how far an eigenvalue $\lambda$ of a signed graph $G^\prime$ can be from the nearest eigenvalue of an induced subgraph $H$. The problem becomes nontrivial when $\lambda$ is not an eigenvalue of $H$. The authors are successful in estimating the distance in terms of eigenvectors and structural parameters related to vertex degrees. From the well-known interlacing theorem, it follows that, if the multiplicity of $\lambda$ in the spectrum of $G^\prime$ is $k$, then it belongs to the spectrum of every induced subgraph obtained by deleting at most $k-1$ vertices. So it is interesting to consider this problem when $H^\prime$ is obtained by deleting $k$ vertices of $G$. The authors also consider the case in which $\lambda$ is a simple eigenvalue of $H^\prime$ but $H^\prime$ is not necessarily a vertex-deleted subgraph. Further, they probe the case when $\lambda$ is the spectral radius of an unsigned graph. They also indicate the scope for further research in detail.
0 references
signed graph
0 references
adjacency matrix
0 references
eigenvalue
0 references
eigenspace
0 references
upper bound
0 references
spectral distance
0 references