TheoremBase

The Real Matrices with the Operator-Norm Distance Form a Complete Metric Space with Entrywise Convergence

For all sizes, the operator-norm distance is a complete metric on the real matrices, bounding the differences of entries, of norms and of matrix-vector products, and convergence and the Cauchy property in it are entrywise.

Statement

In the setting of Routine Facts Used Without Citation: Arithmetic, Maps, Predicativity and Indices, for r,s∈Nr,s\in\mathbb{N} let Mr×s(R)\mathcal{M}_{r\times s}(\mathbb{R}) be the set of real matrices, with entries AijA_{ij}, the matrix-vector product AxAx, the operator norm ∥A∥\lVert A\rVert and the distance dr×sd_{r\times s}; let ∥x∥\lVert x\rVert be the Euclidean norm on Rq\mathbb{R}^{q} for q∈Nq\in\mathbb{N}, with the difference x−yx-y of x,y∈Rqx,y\in\mathbb{R}^{q}, and dRd_{\mathbb{R}} the distance on R\mathbb{R} of Metric Spaces: Balls, the Metric Topology, and the Real and Complex Numbers as Metric Spaces §real-line. For a sequence (A(k))(A^{(k)}) in Mr×s(R)\mathcal{M}_{r\times s}(\mathbb{R}), i∈[r]i\in[r] and j∈[s]j\in[s], (Aij(k))(A^{(k)}_{ij}) is the sequence k↦Aij(k)k\mapsto A^{(k)}_{ij} in R\mathbb{R} of The Sequences k ↦ f_k(p) of a Sequence in ℝ^I, and the Map of Their Limits §sections at the point (i,j)(i,j), as Mr×s(R)=R[r]×[s]\mathcal{M}_{r\times s}(\mathbb{R})=\mathbb{R}^{[r]\times[s]} by Real Matrices: the Set of Real m x n Matrices with Their Entrywise Operations §matrices. Convergence and Cauchy sequences are those of a metric space.

For all r,s∈Nr,s\in\mathbb{N}, dr×sd_{r\times s} is a metric on Mr×s(R)\mathcal{M}_{r\times s}(\mathbb{R}), so that (Mr×s(R),dr×s)(\mathcal{M}_{r\times s}(\mathbb{R}),d_{r\times s}) is a metric space.

For all r,s∈Nr,s\in\mathbb{N}, A,B∈Mr×s(R)A,B\in\mathcal{M}_{r\times s}(\mathbb{R}), i∈[r]i\in[r] and j∈[s]j\in[s], ∣Aij−Bij∣≤dr×s(A,B)|A_{ij}-B_{ij}|\le d_{r\times s}(A,B).

For all r,s∈Nr,s\in\mathbb{N} and A,B∈Mr×s(R)A,B\in\mathcal{M}_{r\times s}(\mathbb{R}), ∣∥A∥−∥B∥∣≤dr×s(A,B)\bigl|\lVert A\rVert-\lVert B\rVert\bigr|\le d_{r\times s}(A,B).

For all r,s∈Nr,s\in\mathbb{N}, A,B∈Mr×s(R)A,B\in\mathcal{M}_{r\times s}(\mathbb{R}) and x∈Rsx\in\mathbb{R}^{s}, ∥Ax−Bx∥≤dr×s(A,B) ∥x∥\lVert Ax-Bx\rVert\le d_{r\times s}(A,B)\,\lVert x\rVert.

For all r,s∈Nr,s\in\mathbb{N}, every sequence (A(k))(A^{(k)}) in Mr×s(R)\mathcal{M}_{r\times s}(\mathbb{R}) and every A∈Mr×s(R)A\in\mathcal{M}_{r\times s}(\mathbb{R}), (A(k))(A^{(k)}) converges to AA in (Mr×s(R),dr×s)(\mathcal{M}_{r\times s}(\mathbb{R}),d_{r\times s}) if and only if (Aij(k))(A^{(k)}_{ij}) converges to AijA_{ij} in (R,dR)(\mathbb{R},d_{\mathbb{R}}) for all i∈[r]i\in[r] and j∈[s]j\in[s].

For all r,s∈Nr,s\in\mathbb{N} and every sequence (A(k))(A^{(k)}) in Mr×s(R)\mathcal{M}_{r\times s}(\mathbb{R}), (A(k))(A^{(k)}) is a Cauchy sequence in (Mr×s(R),dr×s)(\mathcal{M}_{r\times s}(\mathbb{R}),d_{r\times s}) if and only if (Aij(k))(A^{(k)}_{ij}) is a Cauchy sequence in (R,dR)(\mathbb{R},d_{\mathbb{R}}) for all i∈[r]i\in[r] and j∈[s]j\in[s].

For all r,s∈Nr,s\in\mathbb{N}, (Mr×s(R),dr×s)(\mathcal{M}_{r\times s}(\mathbb{R}),d_{r\times s}) is complete.

Proofs

Log in to submit a proof.

Loading...

Citations

Loading…

Dependencies

Loading…

Related

0 relations

Curated associations between results. These are editable and subjective — they do not replace the dependency graph, which is derived from the references in the text.

No relations recorded yet.

Comments

Log in to comment.

Loading…