Skip to main content

All Questions

Filter by
Sorted by
Tagged with
3 votes
1 answer
200 views

Matrix-tree theorem for inverse matrices

Let $L$ be the Laplacian of a directed weighted graph on $n$ nodes, e.g., for $n=4$: $$ L = \left(\begin{array}{cccc} w_{1,1}+w_{1,2}+w_{1,3}+w_{1,4} & -w_{1,2} & -w_{1,3} & -w_{1,4}\\ ...
Federico Poloni's user avatar
2 votes
1 answer
140 views

Invertibility of message passing with invertible parametrization

Consider the message passing framework defined by, $$f(\boldsymbol{x}_i)= \boldsymbol{x}_i + \sum_{j \neq i} (\boldsymbol{x}_i -\boldsymbol{x}_j) g(\|\boldsymbol{x}_i -\boldsymbol{x}_j\|^2),$$ for $i=...
PonderingPolynomial's user avatar
2 votes
0 answers
97 views

Bound on the magnitude of the entries of the Laplacian pseudo-inverse

Let $L$ be the laplacian matrix of a connected graph $G$ with real positive weights and $N$ vertices, or that can be assumed to have binary weights for simplicity.My goal is to bound $\Vert L^+\Vert_{\...
sd24's user avatar
  • 21
3 votes
1 answer
1k views

Calculating a generalized inverse (Moore–Penrose pseudoinverse)

I am considering a graph with $n$ edges with the following nicely structured adjacency matrix: \begin{equation} A_n= \begin{pmatrix} 0 & 0 & 0 &\cdots & 0 & 0 & 1\\ 0 & 0 &...
MthQ's user avatar
  • 41