\(A = CMR\) Factorization
Factor a matrix as \(A = CMR\)1,
where \(M = W^{-1}\) and \(R\) is the matrix of linearly independent rows of \(A\).
\(W\) is the intersection submatrix2 at the intersection of pivot columns and rows.
The \(A = CMR\) factorization expresses a matrix as:
\[ A = CMR, \qquad M = W^{-1} \]
where:
- \(C\) is an \(m \times r\) matrix of linearly independent columns from \(A\)
- \(R\) is an \(r \times n\) matrix of linearly independent rows from \(A\)
- \(W\) is the \(r \times r\) invertible submatrix at the intersection3, and \(M = W^{-1}\) is its inverse
- \(r = \operatorname{rank}(A)\)
Construction:
- Identify pivot positions in \(\operatorname{RREF}(A)\)
- Extract pivot columns from \(A\) to form \(C\)
- Extract pivot rows from \(A\) to form \(R\)
- \(W\) is the submatrix of \(A\) at the intersection of these rows/columns; then \(M = W^{-1}\)
Applications:
- Expresses \(A\) in terms of its essential structure
- Generalizes LU factorization4
- Useful in numerical linear algebra
Enter dimensions and click Generate Matrix to create the input grid.
1 — Compute
Computes the rank-revealing \(A = CMR\) factorization of the matrix.
Resets all dimensions, matrix entries, and results.